NVIDIA RTX A1000 vs NVIDIA Tesla M60 Comparison

The Verdict

The benchmark database puts the NVIDIA RTX A1000 clearly ahead of the NVIDIA Tesla M60 in every recorded head-to-head test. The RTX A1000 wins 2 out of 2 comparisons, with a 76.5% lead in Geekbench OpenCL and a 57.5% lead in Geekbench Vulkan. The average benchmark score for the RTX A1000 is 34207, placing it in the 79th percentile of all GPUs, while the Tesla M60 averages 30490 and sits in the 75th percentile.

The RTX A1000 is the pick for anyone running compute workloads that use OpenCL or Vulkan, and its advantage is not marginal. The delta between the two cards in OpenCL is enormous, and the Vulkan gap is also substantial. The Tesla M60, by contrast, is an end-of-life product from 2015, and the data shows it simply cannot keep pace with a modern workstation card.

The RTX A1000 also brings a much lower power draw of 50 W versus 300 W for the Tesla M60, a single-slot form factor versus dual-slot, and no external power connector requirement. The Tesla M60 needs one 8-pin connector and a suggested 700 W power supply, whereas the RTX A1000 suggests only 250 W. For deployment flexibility, the RTX A1000 is the obvious choice.

The Tesla M60 still has some strengths in raw throughput metrics like pixel rate and texture rate, which we will cover in the Architecture Differences section, but those do not translate into wins in the recorded application benchmarks. The data is unambiguous: the RTX A1000 is the superior card for the workloads represented in the database.

Architecture Differences

The two cards come from very different eras of NVIDIA's design roadmap. The RTX A1000 uses the GA107 chip built on Ampere architecture, fabricated on an 8 nm process at Samsung. It packs 8,700 million transistors into a 200 mm² die, giving a transistor density of 43.5M per mm². The Tesla M60 uses the GM204 chip on Maxwell 2.0 architecture, built on a 28 nm process at TSMC, with 5,200 million transistors on a 398 mm² die, for a density of only 13.1M per mm². The RTX A1000 crams almost twice the transistors into half the silicon area.

Memory configurations are similar in capacity but different in technology. Both cards have 8 GB of VRAM. The RTX A1000 uses GDDR6 on a 128-bit bus with 192.0 GB/s of bandwidth. The Tesla M60 uses GDDR5 on a 256-bit bus with 160.4 GB/s. Despite the narrower bus, the RTX A1000 achieves higher memory bandwidth thanks to faster GDDR6 memory running at 1500 MHz (12 Gbps effective) versus 1253 MHz (5 Gbps effective) on the Tesla.

The compute resources tell an interesting story. The RTX A1000 has 2304 shading units, 72 TMUs, and 32 ROPs. The Tesla M60 has 2048 shading units, 128 TMUs, and 64 ROPs. So the Tesla M60 has more texture units and ROPs, which explains its higher pixel rate of 75.39 GPixel/s and texture rate of 150.8 GTexel/s, compared to 46.78 GPixel/s and 105.3 GTexel/s on the RTX A1000. However, the RTX A1000 has higher raw FP32 throughput at 6.737 TFLOPS versus 4.825 TFLOPS, and it adds features the Tesla M60 lacks entirely: 18 RT cores and 72 tensor cores. The Tesla M60 has no RT cores and no tensor cores at all.

Feature support also diverges. The RTX A1000 supports DirectX 12 Ultimate (12_2), while the Tesla M60 tops out at DirectX 12 (12_1). Both support OpenGL 4.6 and Vulkan 1.4. The RTX A1000 has display outputs (4x mini-DisplayPort 1.4a), while the Tesla M60 has no display outputs at all, as it is a compute-focused accelerator. The bus interface differs too: PCIe 4.0 x8 on the RTX A1000 versus PCIe 3.0 x16 on the Tesla M60.

Clock speeds are notably different. The RTX A1000 runs at 727 MHz base and 1462 MHz boost. The Tesla M60 runs at 557 MHz base and 1178 MHz boost. The RTX A1000 clocks higher despite drawing far less power. The production status also differs: the RTX A1000 is active and was released in 2024, while the Tesla M60 is end-of-life and was released in 2015. The Tesla M60's predecessor is Tesla Kepler and its successor is Tesla Pascal. The RTX A1000's predecessor is Quadro Turing and its successor is Workstation Ada.

Head-to-Head Benchmarks

The database records two direct comparisons between these cards, and the RTX A1000 wins both by wide margins.

In Geekbench OpenCL, the RTX A1000 scores 52078 against 29506 for the Tesla M60. That is a delta of 76.5%. This is the larger of the two gaps, and it reflects the RTX A1000's modern architecture, higher FP32 throughput, and faster memory bandwidth. The Tesla M60, despite having more texture units and ROPs, cannot overcome the generational deficit in compute efficiency. A 76.5% lead in a general compute benchmark is decisive.

In Geekbench Vulkan, the RTX A1000 scores 49574 against 31473 for the Tesla M60, a delta of 57.5%. This gap is smaller than the OpenCL gap but still dominant. Vulkan workloads tend to exercise a mix of compute and graphics paths, and the RTX A1000's RT cores and tensor cores may play a role here, though the benchmark data does not break down per-feature contributions. What the data shows is a consistent, large advantage for the newer card across two different API families.

The average benchmark score for the RTX A1000 is 34207, which puts it 12.2% above the Tesla M60's 30490 average. In terms of percentile ranking, the RTX A1000 sits at the 79th percentile of all GPUs, while the Tesla M60 sits at the 75th. The nearest rivals for the RTX A1000 are the NVIDIA RTX A2000 12 GB at 34154 (0.2% behind), the AMD Radeon RX 560 XT at 34133 (0.2% behind), the NVIDIA TITAN V at 34355 (0.4% ahead), and the AMD Radeon RX 480 at 33997 (0.6% behind). The Tesla M60's nearest rivals are the NVIDIA CMP 70HX at 30476 (0.0% delta), the AMD Radeon RX 6700 at 30433 (0.2% ahead), the AMD Radeon RX 6800 at 30095 (1.3% ahead), and the NVIDIA GeForce RTX 3070 Ti at 29945 (1.8% ahead). These rival comparisons show that both cards sit in competitive neighborhoods for their respective performance tiers, but the RTX A1000's tier is simply higher.

FAQ

Q: Which card has higher raw compute throughput?

A: The RTX A1000 delivers 6.737 TFLOPS of FP32 compute, while the Tesla M60 delivers 4.825 TFLOPS. The RTX A1000 also supports FP16 at 6.737 TFLOPS (1:1 ratio), while the Tesla M60 has no recorded FP16 capability.

Q: How do the memory systems compare?

A: Both cards have 8 GB of VRAM. The RTX A1000 uses GDDR6 on a 128-bit bus with 192.0 GB/s bandwidth. The Tesla M60 uses GDDR5 on a 256-bit bus with 160.4 GB/s bandwidth. The RTX A1000 achieves higher bandwidth despite the narrower bus.

Q: Does the Tesla M60 have any advantages in the recorded data?

A: Yes, in pixel rate and texture rate. The Tesla M60 achieves 75.39 GPixel/s and 150.8 GTexel/s, while the RTX A1000 achieves 46.78 GPixel/s and 105.3 GTexel/s. These advantages do not translate into wins in the recorded OpenCL or Vulkan benchmarks.

Q: What are the power requirements for each card?

A: The RTX A1000 has a TDP of 50 W, requires no power connectors, and suggests a 250 W power supply. The Tesla M60 has a TDP of 300 W, requires one 8-pin connector, and suggests a 700 W power supply.

Q: Which card supports ray tracing and tensor operations?

A: Only the RTX A1000. It has 18 RT cores and 72 tensor cores. The Tesla M60 has none.

Q: Are both cards still in production?

A: No. The RTX A1000 is listed as active, with a release date in 2024. The Tesla M60 is end-of-life, with a release date in 2015.

Where Each One Wins

The RTX A1000 wins in compute-focused workloads. Its 76.5% lead in Geekbench OpenCL and 57.5% lead in Geekbench Vulkan make it the clear choice for general-purpose GPU compute, particularly for applications that leverage OpenCL or Vulkan. The RTX A1000 also wins decisively on power efficiency: 50 W TDP versus 300 W, no power connector versus an 8-pin, and a suggested 250 W power supply versus 700 W. For dense deployments where power and cooling are constrained, the RTX A1000 is the only reasonable option between the two. Its single-slot form factor and 163 mm length also make it far easier to fit into space-constrained systems compared to the Tesla M60's dual-slot, 267 mm design.

The RTX A1000 additionally wins on modern feature support. It has RT cores and tensor cores, DirectX 12 Ultimate support, and display outputs. The Tesla M60 has none of these. For any workload that touches ray tracing, tensor operations, or requires a display output, the RTX A1000 is the only functional choice. The PCIe 4.0 x8 interface on the RTX A1000 also provides higher per-lane bandwidth than the Tesla M60's PCIe 3.0 x16, which matters for data transfer-bound workloads.

The Tesla M60's wins are limited to specific rasterization throughput metrics. Its 75.39 GPixel/s pixel rate is 61% higher than the RTX A1000's 46.78 GPixel/s, and its 150.8 GTexel/s texture rate is 43% higher than the RTX A1000's 105.3 GTexel/s. These figures suggest the Tesla M60 could theoretically excel in fill-rate-bound graphics tasks, but the recorded benchmark data shows no such advantage materializing in OpenCL or Vulkan. The Tesla M60 also has more TMUs (128 versus 72) and more ROPs (64 versus 32), but these do not compensate for the RTX A1000's advantages in shading units, clock speed, memory bandwidth, and architecture efficiency.

In summary, the RTX A1000 is the superior card across every recorded benchmark and most architectural metrics. The Tesla M60 retains niche fill-rate advantages that are not reflected in application-level performance. For any practical workload represented in the database, the RTX A1000 is the card to choose. The Tesla M60, as an end-of-life product from 2015, is best relegated to legacy deployments where its specific pixel and texture throughput characteristics are already integrated into an existing pipeline.