GPU Comparison

The NVIDIA Quadro P2000 and AMD Radeon HD 8790M represent two distinct eras of mobile and workstation graphics. The data shows a decisive performance gap, with the Quadro P2000 leading across every benchmark where both are tested. This analysis breaks down the exact figures, architectural foundations, and the practical implications of these results based solely on the provided specifications.

Head-to-Head Benchmarks

The only two benchmark tests shared between these GPUs are Geekbench OpenCL and Geekbench Vulkan, and the NVIDIA Quadro P2000 wins both by a substantial margin. In Geekbench OpenCL, the Quadro P2000 scores 20,125 points against the AMD Radeon HD 8790M's 5,017 points. This represents a 301.1% advantage for the NVIDIA part, meaning the Quadro P2000 delivers roughly four times the compute performance in this API. The margin in Geekbench Vulkan is similarly lopsided: the Quadro P2000 scores 23,566 points versus 6,365 points for the AMD part, a 270.2% lead. In both tests, the Quadro P2000's output is more than triple that of the Radeon HD 8790M.

The average benchmark score across all tests tells a consistent story. The Quadro P2000 averages 6,049 points, while the Radeon HD 8790M averages 5,691 points, a gap of roughly 6.3%. However, this average is heavily influenced by the limited test set for the AMD part, which only includes the two Geekbench entries. The head-to-head comparisons are more telling: the Quadro P2000 wins 2 out of 2 shared tests, with no wins for the AMD part. The delta percentages in these direct comparisons are the largest in the dataset, far exceeding the margins seen in the nearest-rival comparisons for either card.

Looking at the broader benchmark context, the Quadro P2000 has a much more extensive test profile. Its Passmark G3D score is 6,956, and its Passmark G2D score is 626. It also posts scores in DirectX 9 (124), DirectX 10 (34), DirectX 11 (47), and DirectX 12 (28), plus a Passmark GPU Compute score of 2,933. The Radeon HD 8790M has no Passmark entries listed, so its capabilities in those legacy and compute workloads cannot be quantified here. The Quadro P2000's percentile rank of 35 versus the AMD part's rank of 33 indicates both sit in the lower-middle range of all GPUs, but the NVIDIA card is positioned slightly higher despite its much stronger head-to-head performance.

Architecture Differences

The architectural gap between these two GPUs is generational and profound. The Quadro P2000 is built on NVIDIA's Pascal architecture, utilizing the GP106 chip fabricated on a 16 nm TSMC process. The Radeon HD 8790M uses AMD's GCN 1.0 architecture with the Mars chip on a 28 nm TSMC process. The process node difference alone is significant: 16 nm versus 28 nm, which directly impacts transistor density. The Quadro P2000 packs 4,400 million transistors into a 200 mm² die, yielding a density of 22.0M transistors per mm². The Radeon HD 8790M contains just 950 million transistors on a 77 mm² die, with a density of 12.3M transistors per mm². This means the NVIDIA chip has nearly 4.6 times as many transistors on a die that is only 2.6 times larger.

Core configuration differences further explain the performance gap. The Quadro P2000 features 1,024 shading units, 64 texture mapping units (TMUs), and 40 render output units (ROPs). The Radeon HD 8790M has 384 shading units, 24 TMUs, and only 8 ROPs. The NVIDIA part therefore has 2.7 times the shading units, 2.7 times the TMUs, and 5 times the ROPs. These figures translate directly into throughput: the Quadro P2000 achieves a pixel rate of 59.20 GPixel/s and a texture rate of 94.72 GTexel/s, versus the AMD part's 7.200 GPixel/s and 21.60 GTexel/s. The FP32 compute rating is 3.031 TFLOPS for the NVIDIA card, compared to 691.2 GFLOPS for the AMD card, a 4.4 times advantage. Notably, the Quadro P2000 also lists an FP16 rating of 47.36 GFLOPS (at a 1:64 ratio), while the Radeon HD 8790M has no FP16 data provided.

Memory architecture is another major differentiator. The Quadro P2000 ships with 5 GB of GDDR5 on a 160-bit bus, producing 140.2 GB/s of bandwidth. The Radeon HD 8790M offers 2 GB of GDDR5 on a 128-bit bus, with 64.00 GB/s of bandwidth. The NVIDIA card has 2.5 times the memory capacity and more than double the memory bandwidth. Clock speeds also favor the newer part: the Quadro P2000 runs at a base of 1076 MHz with a boost of 1480 MHz, while the Radeon HD 8790M operates at 850 MHz base and 900 MHz boost. The memory clock difference is stark, the Quadro P2000 runs its GDDR5 at 1752 MHz (7 Gbps effective), while the AMD part runs at 1000 MHz (4 Gbps effective). These architecture and memory differences compound to produce the overwhelming benchmark lead.

The Verdict

The data is unambiguous: the NVIDIA Quadro P2000 is the superior performer in every measurable metric common to both cards. For any workload that relies on OpenCL or Vulkan compute, the Quadro P2000 offers a 270% to 301% performance advantage. Its higher shading unit count, TMU count, ROP count, and memory bandwidth make it the clear choice for tasks like rendering, compute offload, or any GPU-accelerated application that can leverage these resources. The 5 GB of memory versus 2 GB is also a practical advantage for larger datasets or higher-resolution textures.

The Radeon HD 8790M, by contrast, is an older mobile part with a much smaller resource pool. Its 384 shading units and 8 ROPs limit its pixel throughput to 7.200 GPixel/s, which is less than one-eighth of the Quadro P2000's rate. Its 64.00 GB/s bandwidth is less than half of the NVIDIA card's, and its total FP32 throughput of 691.2 GFLOPS is a fraction of the 3.031 TFLOPS offered by the Quadro P2000. Users should pick the Radeon HD 8790M only if they require its specific MXM-A (3.0) form factor or if their workload is constrained by the AMD part's portable-device-dependent display outputs, but from a pure performance standpoint, the data offers no reason to choose it over the Quadro P2000.

For professional users who need a single-slot card with no power connectors and a 75 W TDP, the Quadro P2000 provides that capability while also delivering top-tier performance within its class. The Radeon HD 8790M's lack of TDP data and its MXM Module slot width indicate it is designed for laptop or proprietary systems, not standard workstation slots. The Quadro P2000's PCIe 3.0 x16 interface and four DisplayPort 1.4a outputs make it a more versatile and modern solution. The benchmark data, the architectural resources, and the interface options all point to the same conclusion: the Quadro P2000 is the only rational choice where performance matters.

Specification Differences

| Specification | NVIDIA Quadro P2000 | AMD Radeon HD 8790M |

|---|---|---|

| Architecture | Pascal | GCN 1.0 |

| Process Node | 16 nm | 28 nm |

| Transistors | 4,400 million | 950 million |

| Die Size | 200 mm² | 77 mm² |

| Transistor Density | 22.0M / mm² | 12.3M / mm² |

| Base Clock | 1076 MHz | 850 MHz |

| Boost Clock | 1480 MHz | 900 MHz |

| Memory Clock | 1752 MHz (7 Gbps effective) | 1000 MHz (4 Gbps effective) |

| Memory Size | 5 GB | 2 GB |

| Memory Bus Width | 160 bit | 128 bit |

| Memory Bandwidth | 140.2 GB/s | 64.00 GB/s |

| Shading Units | 1024 | 384 |

| TMUs | 64 | 24 |

| ROPs | 40 | 8 |

| Pixel Rate | 59.20 GPixel/s | 7.200 GPixel/s |

| Texture Rate | 94.72 GTexel/s | 21.60 GTexel/s |

| FP32 | 3.031 TFLOPS | 691.2 GFLOPS |

| FP16 | 47.36 GFLOPS (1:64) |, |

| TDP | 75 W |, |

| Slot Width | Single-slot | MXM Module |

| Bus Interface | PCIe 3.0 x16 | MXM-A (3.0) |

| Display Outputs | 4x DisplayPort 1.4a | Portable Device Dependent |

| Vulkan Version | 1.4 | 1.2.170 |

| Release Date | 2017-02-05 | 2013-03-31 |

| Predecessor | Quadro Maxwell | London |

| Successor | Quadro Volta | Gem System |

FAQ

Q: Which GPU wins the Geekbench OpenCL benchmark?

A: The NVIDIA Quadro P2000 wins decisively with a score of 20,125 versus the AMD Radeon HD 8790M's 5,017, a 301.1% advantage.

Q: What is the difference in memory bandwidth between the two cards?

A: The Quadro P2000 offers 140.2 GB/s of bandwidth, while the Radeon HD 8790M provides 64.00 GB/s, the NVIDIA part has more than double the bandwidth.

Q: How do the shading unit counts compare?

A: The Quadro P2000 has 1,024 shading units, compared to the Radeon HD 8790M's 384 shading units, giving the NVIDIA card 2.7 times as many.

Q: Which card has a higher transistor density?

A: The Quadro P2000 has a transistor density of 22.0M per mm², versus the Radeon HD 8790M's 12.3M per mm², due to its smaller 16 nm process node versus 28 nm.

Q: Are there any benchmarks where the AMD Radeon HD 8790M wins?

A: No. In the head-to-head dataset, the Quadro P2000 wins 2 out of 2 shared tests, and the AMD part has 0 wins.

Q: What is the average benchmark score for each GPU?

A: The Quadro P2000 averages 6,049 points, while the Radeon HD 8790M averages 5,691 points, a difference of roughly 6.3% in favor of the NVIDIA card.