NVIDIA A2 vs NVIDIA GRID M60-1Q Comparison

Head-to-Head Benchmarks

The only shared benchmark recorded for these two workstation cards is Geekbench Vulkan, and the data shows a clear, though not overwhelming, advantage for the NVIDIA A2. The A2 scores 34023 points against 31220 for the GRID M60-1Q. That is a 9% delta in favor of the A2, a meaningful margin in GPU compute workloads but hardly a generational blowout given the architectural gap between the two.

Context from the nearest rivals helps frame this result. The A2's average benchmark score across all recorded tests is 34690, which places it 1% ahead of the NVIDIA TITAN V (34355) and 1.4% ahead of the NVIDIA RTX A1000 (34207). The GRID M60-1Q, by contrast, has an average score of 31220, essentially tied with the NVIDIA Quadro M5000 (31206, 0% delta) and within 0.4% of the GeForce RTX 4070 Ti SUPER (31087). In other words, the A2 sits in a higher performance tier overall, and the Vulkan head-to-head confirms that ordering.

The A2's win is consistent across the two benchmarks it has recorded. Its Geekbench OpenCL score is 35357, which is higher than its Vulkan result, suggesting strong compute throughput in both APIs. The GRID M60-1Q has no OpenCL record in the database, so the only direct comparison is the Vulkan test, where the A2 leads by the 9% margin noted above. The head-to-head tally is therefore 1 win for the A2 and 0 for the GRID M60-1Q, a clean sweep but based on a single shared test.

Interpreting the 9% gap requires attention to the underlying hardware differences. The A2 is built on a much newer architecture and process, which compensates for its lower raw shading unit count. The GRID M60-1Q actually fields more shading units (2048 vs 1280), more texture mapping units (128 vs 40), and more render output units (64 vs 32). Its theoretical peak FP32 throughput is also higher at 4.825 TFLOPS versus 4.531 TFLOPS. Yet the A2 still wins the Vulkan test. This indicates that the A2's Ampere architecture extracts more real-world performance per FLOP, likely due to newer instruction efficiency and better driver optimization for modern APIs. The GRID M60-1Q's Maxwell 2.0 design, while still capable, does not translate its higher theoretical numbers into a benchmark victory.

Another factor is memory configuration. The A2 carries 16 GB of GDDR6 on a 128-bit bus, delivering 200.1 GB/s of bandwidth. The GRID M60-1Q has only 1024 MB of GDDR5 on a 256-bit bus, with 160.4 GB/s of bandwidth. Despite the narrower bus, the A2's faster memory (12.5 Gbps effective vs 5 Gbps effective) yields 25% more bandwidth. That bandwidth advantage likely helps in Vulkan workloads that are memory-sensitive, and the 16x capacity difference (16 GB vs 1 GB) is enormous for any workload that needs to hold large datasets on the card.

The A2 also benefits from newer API support. It lists DirectX 12 Ultimate (12_2), while the GRID M60-1Q only reaches DirectX 12 (12_1). Both cards support OpenGL 4.6 and Vulkan 1.4, so the gap is limited to the DirectX feature level. For Vulkan specifically, both are at the same API version, making the A2's 9% lead purely a matter of execution efficiency and memory throughput rather than API capability.

The Verdict

From the recorded data, the NVIDIA A2 is the stronger card for modern compute workloads. Its 9% Vulkan lead over the GRID M60-1Q, combined with a higher average benchmark score (34690 vs 31220) and a better percentile ranking (79th vs 76th percentile among all GPUs), makes it the clear choice for anyone running general GPU compute or graphics tasks that leverage Vulkan or OpenCL. The A2 also offers 16 GB of memory versus 1 GB, a 16x capacity advantage that matters for larger models, bigger textures, or multi-tenant virtualization scenarios.

The GRID M60-1Q is not without its own merits. It has higher theoretical FP32 throughput (4.825 TFLOPS vs 4.531 TFLOPS), more texture units (128 vs 40), more ROPs (64 vs 32), and a wider 256-bit memory bus. Its pixel rate is also higher at 75.39 GPixel/s versus 56.64 GPixel/s, and its texture rate of 150.8 GTexel/s far exceeds the A2's 70.80 GTexel/s. In workloads that are purely fill-rate bound or that scale well with raw shading units, the older card could still hold its own. However, the benchmark data does not show any test where the GRID M60-1Q wins. Its single recorded Vulkan score is lower, and it has no OpenCL result to suggest an alternative strength.

For buyers choosing between these two end-of-life products, the A2 is the safer recommendation across the board. It wins the only head-to-head benchmark, has a higher average score, sits in a higher percentile, and offers vastly more memory with faster effective bandwidth. The GRID M60-1Q should only be considered if a workload specifically benefits from its higher texture and pixel rates, or if the 225 W TDP and dual-slot footprint are acceptable trade-offs in a system where the A2's 60 W TDP and single-slot design would be preferable. The A2 also requires no power connectors and suggests a 250 W PSU, whereas the GRID M60-1Q needs a single 8-pin connector and a 550 W PSU. For most users, the power efficiency alone tips the balance.

Architecture Differences

The architectural gulf between these two cards is substantial. The NVIDIA A2 is built on the Ampere architecture using the GA107 chip, fabricated on Samsung's 8 nm process. It packs 8,700 million transistors into a 200 mm² die, yielding a transistor density of 43.5 million per mm². The GRID M60-1Q uses the Maxwell 2.0 architecture with the GM204 chip, fabricated on TSMC's 28 nm process. That older node holds 5,200 million transistors across a larger 398 mm² die, resulting in just 13.1 million transistors per mm². The A2 is therefore more than three times denser, a direct consequence of the 8 nm versus 28 nm process gap.

The A2 includes dedicated ray tracing cores (10) and tensor cores (40), features entirely absent from the GRID M60-1Q, which lists neither. This makes the A2 appropriate for workloads that leverage RTX-accelerated ray tracing or tensor core math, such as AI inference or real-time ray-traced rendering. The GRID M60-1Q has no such acceleration, limiting it to traditional rasterization and compute.

Shading unit counts differ in favor of the older card: 2048 shading units on the GRID M60-1Q versus 1280 on the A2. Texture mapping units follow the same pattern, 128 vs 40, as do render output units, 64 vs 32. This gives the GRID M60-1Q a theoretical advantage in raw fill-rate and texturing work, as reflected in its 75.39 GPixel/s pixel rate and 150.15 GTexel/s texture rate. The A2 counters with 56.56 GPixel/s and 70.80 GTexel/s, lower in both metrics.

The memory subsystems reflect their respective eras. The A2 uses 16 GB of GDDR6 with a 128-bit bus and 200.1 GB/s bandwidth. The GRID M60-1Q uses 1024 MB of GDDR5 with a 256-bit bus and 160.4 GB/s bandwidth. The A2's effective memory clock is 12.5 Gbps versus 5 Gbps on the GRID M60-1Q, which explains how the narrower bus still delivers more bandwidth. The A2 also supports PCIe 4.0 x8, while the GRID M60-1Q is limited to PCIe 3.0 x16. For many workloads the A2's newer bus generation can offset the reduced lane count.

API support differs on the DirectX front. The A2 lists DirectX 12 Ultimate (12_2), while the GRID M60-1Q lists DirectX 12 (12_1). Both support OpenGL 4.6 and Vulkan 1.4. The A2's newer feature set aligns with its Ampere generation, which is also reflected in its generation label "Workstation Ampere (Ax000)" versus "GRID (Mx)" for the older card.

Specification Differences

The following specifications differ between the two cards, based solely on the database records:

• Process node: A2 is 8 nm (Samsung); GRID M60-1Q is 28 nm (TSMC).
• Transistors: A2 has 8,700 million; GRID M60-1Q has 5,200 million.
• Die size: A2 is 200 mm²; GRID M60-1Q is 398 mm².
• Transistor density: A2 is 43.5M per mm²; GRID M60-1Q is 13.1M per mm².
• Base clock: A2 is 1440 MHz; GRID M60-1Q is 557 MHz.
• Boost clock: A2 is 1770 MHz; GRID M60-1Q is 1178 MHz.
• Memory clock: A2 is 1563 MHz (12.5 Gbps effective); GRID M60-1Q is 1253 MHz (5 Gbps effective).
• Memory size: A2 is 16 GB; GRID M60-1Q is 1024 MB.
• Memory type: A2 is GDDR6; GRID M60-1Q is GDDR5.
• Memory bus: A2 is 128 bit; GRID M60-1Q is 256 bit.
• Memory bandwidth: A2 is 200.1 GB/s; GRID M60-1Q is 160.4 GB/s.
• Shading units: A2 has 1280; GRID M60-1Q has 2048.
• Texture mapping units: A2 has 40; GRID M60-1Q has 128.
• Render output units: A2 has 32; GRID M60-1Q has 64.
• Ray tracing cores: A2 has 10; GRID M60-1Q has none.
• Tensor cores: A2 has 40; GRID M60-1Q has none.
• Pixel rate: A2 is 56.64 GPixel/s; GRID M60-1Q is 75.39 GPixel/s.
• Texture rate: A2 is 70.80 GTexel/s; GRID M60-1Q is 150.8 GTexel/s.
• FP32 throughput: A2 is 4.531 TFLOPS; GRID M60-1Q is 4.825 TFLOPS.
• FP16 throughput: A2 is 4.531 TFLOPS (1:1); GRID M60-1Q has no recorded FP16 value.
• TDP: A2 is 60 W; GRID M60-1Q is 225 W.
• Slot width: A2 is single-slot; GRID M60-1Q is dual-slot.
• Power connectors: A2 has none; GRID M60-1Q has 1x 8-pin.
• Suggested PSU: A2 is 250 W; GRID M60-1Q is 550 W.
• Bus interface: A2 is PCIe 4.0 x8; GRID M60-1Q is PCIe 3.0 x16.
• DirectX support: A2 is 12 Ultimate (12_2); GRID M60-1Q is 12 (12_1).
• Length: GRID M60-1Q is 267 mm (10.5 inches); A2 has no recorded length.
• Release date: A2 is 2021-11-09; GRID M60-1Q is 2015-08-29.

FAQ

Q: Which card wins in Vulkan performance?

A: The NVIDIA A2 scores 34023 in Geekbench Vulkan, which is 9% higher than the GRID M60-1Q's 31220.

Q: How much memory does each card have?

A: The A2 has 16 GB of GDDR6 on a 128-bit bus, while the GRID M60-1Q has 1024 MB of GDDR5 on a 256-bit bus.

Q: Does the GRID M60-1Q have ray tracing or tensor cores?

A: No. The GRID M60-1Q lists neither ray tracing cores nor tensor cores, while the A2 has 10 ray tracing cores and 40 tensor cores.

Q: What is the power draw difference?

A: The A2 has a 60 W TDP and requires no power connectors, while the GRID M60-1Q has a 225 W TDP and needs a single 8-pin connector. The suggested PSU is 250 W for the A2 and 550 W for the GRID M60-1Q.

Q: Which card has higher raw FP32 throughput?

A: The GRID M60-1Q has a higher FP32 rating at 4.825 TFLOPS, versus 4.531 TFLOPS for the A2, despite losing the Vulkan benchmark.

Q: Are both cards still in production?

A: No, both are listed as end-of-life in the database. The A2 was released on 2021-11-09, and the GRID M60-1Q on 2015-08-29.