AMD Radeon HD 8970M vs NVIDIA GeForce RTX 2060 SUPER Comparison

Head-to-Head Benchmarks

The recorded data contains one direct benchmark comparison between these two GPUs: the Geekbench OpenCL test. In this test, the NVIDIA GeForce RTX 2060 SUPER scores 76,957, while the AMD Radeon HD 8970M scores 21,237. This yields a delta of 72.4 percent in favor of the NVIDIA card, a substantial margin that places the AMD part at roughly a quarter of the performance level in this compute-oriented workload.

The NVIDIA GeForce RTX 2060 SUPER also holds a decisive advantage in the broader benchmark database. Its average benchmark score of 18,093 places it in the 62nd percentile of all GPUs tracked. The AMD Radeon HD 8970M, by contrast, has an average score of 21,237 from its single recorded Geekbench OpenCL run, yet it sits in the 66th percentile. This apparent discrepancy stems from the different benchmark distributions; the HD 8970M's lone score is measured against a different population of GPUs than the RTX 2060 SUPER's multi-test average.

The RTX 2060 SUPER's nearest rivals in the database include the NVIDIA GeForce RTX 3060 Mobile (average score 18,159, delta of 0.4 percent ahead), the AMD Radeon Pro 5700 (18,189, delta of 0.5 percent ahead), and the AMD Radeon RX 460 (18,373, delta of 1.5 percent ahead). These deltas are all small, indicating the RTX 2060 SUPER sits in a tightly contested performance band. The HD 8970M's nearest rivals tell a different story: the AMD Radeon RX Vega M GL scores 21,153 (0.4 percent behind the HD 8970M), the NVIDIA RTX A4000 Mobile scores 21,379 (0.7 percent ahead), and the NVIDIA Quadro RTX 5000 scores 21,629 (1.8 percent ahead). The HD 8970M is thus clustered with a set of mid-range mobile and workstation parts, while the RTX 2060 SUPER competes with a more recent generation of mainstream GPUs.

Where Each One Wins

The benchmark data is unambiguous: the RTX 2060 SUPER wins the sole head-to-head test, and it does so by a wide margin. The Geekbench OpenCL result of 76,957 versus 21,237 shows the NVIDIA part delivering roughly 3.6 times the raw compute throughput in this OpenCL workload. This is not a narrow victory; it is a generational gap expressed through a single metric.

The HD 8970M's only recorded benchmark, its Geekbench OpenCL score of 21,237, does place it above several nearby rivals in the database. It sits 0.4 percent ahead of the AMD Radeon RX Vega M GL, 1 percent ahead of the NVIDIA GeForce RTX 5050, and 1.8 percent ahead of the NVIDIA Quadro RTX 5000. These are small margins, but they show the HD 8970M can still hold its own against certain newer parts in compute-oriented tasks, at least within the limited scope of this one test.

The RTX 2060 SUPER, however, has a much broader benchmark footprint. Besides its Geekbench OpenCL score, it records results in Geekbench Vulkan (77,402), Passmark G3D (16,462), Passmark GPU Compute (6,721), and several DirectX-specific tests: DirectX 9 at 218, DirectX 11 at 130, DirectX 12 at 61, and DirectX 10 at 111. It also scores 854 in Passmark G2D. These additional data points give the RTX 2060 SUPER a more complete performance profile, covering legacy APIs, modern APIs, and compute workloads. The HD 8970M has no comparable data in these tests, so any assessment of its gaming or legacy API performance must rely solely on the OpenCL result.

Architecture Differences

The two GPUs come from different architectural eras. The AMD Radeon HD 8970M uses the Neptune chip built on GCN 1.0 architecture, fabricated on a 28 nm process at TSMC. It packs 2,800 million transistors on a 212 mm² die, yielding a transistor density of 13.2 million per square millimeter. The NVIDIA GeForce RTX 2060 SUPER uses the TU106 chip based on Turing architecture, also fabricated at TSMC but on a 12 nm process. It contains 10,800 million transistors on a 445 mm² die, for a density of 24.3 million per square millimeter. The RTX 2060 SUPER thus has nearly four times the transistor count and nearly double the die area, with a higher transistor density reflecting the more advanced process node.

Memory configurations differ substantially. The HD 8970M has 4 GB of GDDR5 on a 256-bit bus, with memory clocked at 1200 MHz (4.8 Gbps effective) and bandwidth of 153.6 GB/s. The RTX 2060 SUPER has 8 GB of GDDR6 on the same 256-bit bus width, but memory runs at 1750 MHz (14 Gbps effective), delivering 448.0 GB/s of bandwidth. That is nearly three times the memory bandwidth, which matters for high-resolution textures and compute workloads that are bandwidth-sensitive.

Compute resources also scale sharply. The HD 8970M has 1,280 shading units, 80 texture mapping units, and 32 ROPs. The RTX 2060 SUPER has 2,176 shading units, 136 TMUs, and 64 ROPs. Its pixel rate is 105.6 GPixel/s versus 28.80 GPixel/s for the HD 8970M, and its texture rate is 224.4 GTexel/s versus 72.00 GTexel/s. Floating-point performance follows the same pattern: the RTX 2060 SUPER delivers 7.181 TFLOPS in FP32, while the HD 8970M delivers 2.304 TFLOPS. The RTX 2060 SUPER also supports FP16 at 14.36 TFLOPS (2:1 ratio), a feature the HD 8970M does not expose in the database.

The RTX 2060 SUPER includes dedicated hardware absent from the older part: 34 ray tracing cores and 272 tensor cores. These enable DirectX 12 Ultimate (12_2) support, whereas the HD 8970M is limited to DirectX 12 (11_1). The RTX 2060 SUPER also supports Vulkan 1.4, while the HD 8970M supports Vulkan 1.2.170. Both cards support OpenGL 4.6 and use PCIe 3.0 x16 interfaces.

Power and physical characteristics differ as well. The HD 8970M has a TDP of 100 W and uses an MXM module form factor, typical for laptops. The RTX 2060 SUPER has a TDP of 175 W, is a dual-slot card, requires a single 8-pin power connector, and has a suggested power supply of 450 W. Its dimensions are 229 mm in length, 113 mm in height, and 35 mm in width. The HD 8970M's display outputs are listed as portable device dependent, while the RTX 2060 SUPER offers 1x DVI, 1x HDMI 2.0, 2x DisplayPort 1.4a, and 1x USB Type-C.

Release dates are separated by roughly six years: the HD 8970M launched on May 13, 2013, and the RTX 2060 SUPER on July 8, 2019. Both are now end-of-life products. The HD 8970M's predecessor is London and its successor is Gem System; the RTX 2060 SUPER's predecessor is GeForce 10 and its successor is GeForce 30.

The Verdict

The data points to one clear conclusion: the NVIDIA GeForce RTX 2060 SUPER is the faster GPU by a wide margin. Its Geekbench OpenCL score of 76,957 is 72.4 percent higher than the HD 8970M's 21,237. It has more shading units, more memory, more bandwidth, higher clock speeds, and dedicated ray tracing and tensor hardware. Its average benchmark score of 18,093 places it in the 62nd percentile of all GPUs, which is lower than the HD 8970M's 66th percentile, but that percentile is computed against different reference sets and does not change the head-to-head result.

For users choosing between these two, the RTX 2060 SUPER is the appropriate pick for modern workloads: DirectX 12 Ultimate titles, ray-traced games, Vulkan applications, and compute tasks that leverage FP16 or tensor cores. Its 8 GB of GDDR6 memory and 448.0 GB/s bandwidth also make it better suited to higher-resolution textures and larger datasets.

The HD 8970M remains relevant only in legacy contexts. Its 4 GB of GDDR5 memory and 153.6 GB/s bandwidth were respectable for its era, and its OpenCL score of 21,237 shows it can still outperform a few newer parts like the AMD Radeon RX Vega M GL and NVIDIA GeForce RTX 5050 in that specific test. But its lack of ray tracing, tensor cores, and modern API support (DirectX 12_1 at best) limits its usefulness in current software. The RTX 2060 SUPER is the clear choice for anyone prioritizing performance and feature support, while the HD 8970M is only sensible for systems that require its MXM form factor and low 100 W TDP.

FAQ

Q: Which GPU is faster in the recorded head-to-head benchmark?

A: The NVIDIA GeForce RTX 2060 SUPER scores 76,957 in Geekbench OpenCL versus 21,237 for the AMD Radeon HD 8970M, a delta of 72.4 percent in favor of the NVIDIA card.

Q: Does the AMD Radeon HD 8970M outperform any of its nearest rivals?

A: Yes, its Geekbench OpenCL score of 21,237 is 0.4 percent ahead of the AMD Radeon RX Vega M GL (21,153) and 1 percent ahead of the NVIDIA GeForce RTX 5050 (21,035).

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

A: The RTX 2060 SUPER has 448.0 GB/s of bandwidth from 8 GB of GDDR6 on a 256-bit bus, while the HD 8970M has 153.6 GB/s from 4 GB of GDDR5 on the same bus width.

Q: Does the RTX 2060 SUPER support ray tracing?

A: Yes, it has 34 ray tracing cores and 272 tensor cores, which support DirectX 12 Ultimate (12_2). The HD 8970M has no such hardware and is limited to DirectX 12 (11_1).

Q: How do the power requirements compare?

A: The HD 8970M has a TDP of 100 W and uses an MXM module, while the RTX 2060 SUPER has a TDP of 175 W, is dual-slot, and requires a 450 W suggested power supply.

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

A: The RTX 2060 SUPER has an average benchmark score of 18,093, while the HD 8970M's average is 21,237 from its single recorded test. The RTX 2060 SUPER sits in the 62nd percentile of all GPUs, and the HD 8970M sits in the 66th percentile.

Specification Differences

| Specification | AMD Radeon HD 8970M | NVIDIA GeForce RTX 2060 SUPER |

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

| Architecture | GCN 1.0 | Turing |

| Process node | 28 nm | 12 nm |

| Transistors | 2,800 million | 10,800 million |

| Die size | 212 mm² | 445 mm² |

| Transistor density | 13.2M / mm² | 24.3M / mm² |

| Base clock | 850 MHz | 1470 MHz |

| Boost clock | 900 MHz | 1650 MHz |

| Memory clock | 1200 MHz (4.8 Gbps effective) | 1750 MHz (14 Gbps effective) |

| Memory size | 4 GB | 8 GB |

| Memory type | GDDR5 | GDDR6 |

| Memory bandwidth | 153.6 GB/s | 448.0 GB/s |

| Shading units | 1280 | 2176 |

| TMUs | 80 | 136 |

| ROPs | 32 | 64 |

| Ray tracing cores | None | 34 |

| Tensor cores | None | 272 |

| Pixel rate | 28.80 GPixel/s | 105.6 GPixel/s |

| Texture rate | 72.00 GTexel/s | 224.4 GTexel/s |

| FP32 performance | 2.304 TFLOPS | 7.181 TFLOPS |

| FP16 performance | None listed | 14.36 TFLOPS (2:1) |

| TDP | 100 W | 175 W |

| Slot width | MXM Module | Dual-slot |

| Power connectors | None listed | 1x 8-pin |

| Suggested PSU | None listed | 450 W |

| Display outputs | Portable Device Dependent | 1x DVI, 1x HDMI 2.0, 2x DisplayPort 1.4a, 1x USB Type-C |

| DirectX support | 12 (11_1) | 12 Ultimate (12_2) |

| Vulkan support | 1.2.170 | 1.4 |

| Release date | 2013-05-13 | 2019-07-08 |

| Launch MSRP | None listed | 399 USD |