GPU Comparison

The AMD Radeon HD 8690M and NVIDIA Quadro P2000 are separated by four years and two distinct market philosophies, yet their aggregate benchmark scores land them within a hair of each other. The HD 8690M averages 6137 points, while the Quadro P2000 averages 6049 points, a difference of less than 1.5%. However, this near-parity in overall standing is a statistical illusion; the single head-to-head benchmark available reveals a chasm in raw compute performance, with the Quadro P2000 delivering more than three times the OpenCL score of the older AMD part.

Head-to-Head Benchmarks

The sole direct comparison in the data is the Geekbench OpenCL test, and it is a decisive victory for the NVIDIA Quadro P2000. The Quadro scores 20125 points, while the Radeon HD 8690M manages 6137 points. This represents a delta of -69.5% for the AMD card, meaning the Quadro is roughly 228% faster in this specific workload. The margin is enormous, and it underscores the generational leap in compute throughput between a 2013 entry-level mobile chip and a 2017 professional workstation card.

Looking at the wider benchmark landscape, the HD 8690M’s only data point is that same OpenCL score, so its position in the 36th percentile of all GPUs is entirely determined by that result. The Quadro P2000, however, has a much richer benchmark profile. Its Passmark G3D score of 6956 is its strongest showing, while its Passmark DirectX 9 score of 124 indicates strong legacy API performance. The Quadro also posts a Passmark GPU Compute score of 2933, which, while lower than its OpenCL number, still represents a substantial compute workload performance. The HD 8690M has no corresponding DirectX or compute scores in the data, so the comparison is necessarily lopsided.

The aggregate scores tell a curious story. Despite the Quadro’s crushing OpenCL victory, its average benchmark score is dragged down by its poor Passmark DirectX scores (34, 47, and 28 for DX10, DX11, and DX12 respectively). These low numbers pull its average down to 6049, which is actually below the HD 8690M’s single score of 6137. This is a classic case of an average being misleading; the Quadro is clearly the stronger GPU in absolute terms, but its average is penalized by the inclusion of tests that favor older, simpler architectures or that are not representative of its intended professional workload.

The nearest rivals for each card reinforce the idea that both are mid-to-low tier performers. The HD 8690M sits within 1.2% of the AMD Radeon HD 6950, and within 1% of the NVIDIA GeForce MX230 and RTX A400. The Quadro P2000 is nearly identical in average score to the GeForce MX230 and RTX A400, with deltas of -0.5% and -0.5% respectively. This suggests that in aggregate, these two cards compete in the same performance tier, despite the Quadro’s massive single-test advantage.

The Verdict

The data is unambiguous for compute-heavy tasks: the NVIDIA Quadro P2000 is the superior product. Its 20125 OpenCL score dwarfs the 6137 of the AMD Radeon HD 8690M, and its 3.031 TFLOPS FP32 throughput is nearly five times the 624.0 GFLOPS of the AMD part. If your workload relies on OpenCL, CUDA, or general GPU compute, the Quadro is the only rational choice. The HD 8690M is not competitive in this arena.

However, for a user focused purely on legacy DirectX performance, the picture is different. The Quadro’s Passmark DirectX scores are remarkably low, with a DX12 score of just 28 and a DX11 score of 47. These numbers are so low that they suggest the Quadro is poorly optimized for those older APIs, or that the benchmark is not measuring its strengths. The HD 8690M has no DirectX scores in the data, so we cannot directly compare them, but the Quadro’s numbers are objectively weak. If you are running legacy applications that rely on DirectX 9 through 12, the Quadro may disappoint.

The average benchmark scores suggest that for a balanced, multi-test workload, the two cards perform nearly identically. The HD 8690M’s 6137 average is 1.5% higher than the Quadro’s 6049. But this is a hollow victory for AMD, as it is based on a single data point. The Quadro’s average is a more honest representation of its broad capabilities, even if it is dragged down by weak DirectX results. In short, the Quadro is a specialized tool for compute; the HD 8690M is a general-purpose chip from an older era.

Where Each One Wins

The AMD Radeon HD 8690M wins in the narrow category of aggregate benchmark consistency. Its single OpenCL score of 6137 places it at the 36th percentile, and its nearest rivals are all within a 1% to 1.2% delta. This means that within its limited performance envelope, it is a predictable and stable performer. It also has a lower power footprint, with a 28 nm process node and a significantly lower transistor count of 690 million, which may make it a better fit for systems with minimal cooling or power delivery, though the TDP is not listed in the data.

The NVIDIA Quadro P2000 wins decisively in raw compute performance. Its OpenCL score of 20125 is more than triple the HD 8690M’s, and its FP32 throughput of 3.031 TFLOPS is in a different class entirely. For any professional application that leverages GPU acceleration, rendering, simulations, data processing, the Quadro is the clear winner. It also has a more robust feature set for modern APIs, supporting DirectX 12 (12_1) and Vulkan 1.4, whereas the HD 8690M only supports DirectX 12 (11_1) and Vulkan 1.2.170. The Quadro’s 5 GB of VRAM and 140.2 GB/s of bandwidth also provide a massive advantage over the HD 8690M’s 2 GB and 32.00 GB/s, which is critical for large datasets.

FAQ

Q: Which GPU has a higher OpenCL benchmark score?

A: The NVIDIA Quadro P2000 scores 20125 in Geekbench OpenCL, while the AMD Radeon HD 8690M scores 6137. The Quadro is 69.5% higher in this test.

Q: How do their average benchmark scores compare?

A: The AMD Radeon HD 8690M has an average benchmark score of 6137, which is about 1.5% higher than the NVIDIA Quadro P2000’s average of 6049.

Q: Which card has more VRAM?

A: The NVIDIA Quadro P2000 has 5 GB of GDDR5 memory, while the AMD Radeon HD 8690M has 2 GB of GDDR5 memory.

Q: What is the memory bandwidth difference?

A: The Quadro P2000 offers 140.2 GB/s of bandwidth, which is over four times the 32.00 GB/s of the HD 8690M.

Q: Which GPU supports a newer version of DirectX?

A: The NVIDIA Quadro P2000 supports DirectX 12 (12_1), while the AMD Radeon HD 8690M supports DirectX 12 (11_1).

Q: Do either of these cards support Vulkan?

A: Yes, both support Vulkan. The Quadro P2000 supports Vulkan 1.4, while the HD 8690M supports Vulkan 1.2.170.

Architecture Differences

The two GPUs are built on fundamentally different architectures and process nodes. The AMD Radeon HD 8690M uses the Sun chip, based on the GCN 1.0 architecture, fabricated on a 28 nm process at TSMC. It contains 690 million transistors on a 56 mm² die, yielding a transistor density of 12.3M per mm². The NVIDIA Quadro P2000 uses the GP106 chip, based on the Pascal architecture, also fabricated by TSMC but on a 16 nm process. This chip packs 4,400 million transistors onto a 200 mm² die, achieving a density of 22.0M per mm². The Quadro’s newer node allows for a significantly higher transistor count and density.

The compute resources are also vastly different. The HD 8690M has 320 shading units, 20 TMUs, and 8 ROPs. The Quadro P2000 has 1024 shading units, 64 TMUs, and 40 ROPs. This translates to a pixel rate of 7.800 GPixel/s and a texture rate of 19.50 GTexel/s for the AMD card, versus 59.20 GPixel/s and 94.72 GTexel/s for the NVIDIA card. In terms of floating-point performance, the Quadro’s FP32 output is 3.031 TFLOPS, while the HD 8690M is limited to 624.0 GFLOPS. The Quadro also has an FP16 rate of 47.36 GFLOPS (1:64), a feature not listed for the AMD card.

The API support reflects the architectural differences. The Quadro P2000 supports DirectX 12 (12_1) and Vulkan 1.4, while the HD 8690M is limited to DirectX 12 (11_1) and Vulkan 1.2.170. Both cards support OpenGL 4.6. The memory subsystem is another major divergence, with the HD 8690M using a 64-bit bus and the Quadro P2000 using a 160-bit bus.

Specification Differences

The most striking difference is in process technology and die size. The AMD Radeon HD 8690M is built on a 28 nm process with a 56 mm² die, while the NVIDIA Quadro P2000 uses a 16 nm process with a 200 mm² die. This leads to a transistor count of 690 million for AMD versus 4,400 million for NVIDIA.

Clock speeds also differ significantly. The HD 8690M has a base clock of 925 MHz and a boost clock of 975 MHz, with memory clocked at 1000 MHz (4 Gbps effective). The Quadro P2000 has a base clock of 1076 MHz and a boost clock of 1480 MHz, with memory at 1752 MHz (7 Gbps effective). The memory configuration is also different: the HD 8690M has 2 GB of GDDR5 on a 64-bit bus, while the Quadro P2000 has 5 GB of GDDR5 on a 160-bit bus. This yields a bandwidth of 32.00 GB/s for AMD and 140.2 GB/s for NVIDIA.

The bus interface differs, with the HD 8690M using PCIe 3.0 x8 and the Quadro P2000 using PCIe 3.0 x16. The Quadro P2000 has a listed TDP of 75 W, a single-slot form factor, no power connectors, and a suggested PSU of 250 W. It also has dimensions of 196 mm in length and 111 mm in height, and features 4x DisplayPort 1.4a outputs. None of these specifications are listed for the HD 8690M. Finally, the release dates are four years apart, with the HD 8690M launching in 2013 and the Quadro P2000 in 2017.