AMD Radeon 680M vs AMD Radeon HD 7790 Comparison

FAQ

Q: Which GPU has the higher average benchmark score?

A: The AMD Radeon HD 7790 has an average benchmark score of 17,666, while the AMD Radeon 680M has an average of 15,270. The HD 7790 sits at the 61st percentile of all GPUs, whereas the 680M sits at the 57th percentile.

Q: How does the HD 7790 compare to its nearest rivals?

A: The HD 7790 is 0.2% ahead of the NVIDIA GeForce RTX 4060 and 0.4% ahead of the AMD Radeon 780M. It trails the NVIDIA Quadro RTX 4000 by only 0.7%, making it essentially a statistical tie with all four listed rivals.

Q: What is the performance position of the Radeon 680M?

A: The 680M is 0.1% behind both the NVIDIA GeForce GTX 580 and the NVIDIA GeForce RTX 2060. It leads the NVIDIA GeForce RTX 3050 OEM by 0.5% and the AMD Radeon RX 7600 by 0.7%.

Q: What are the memory configurations of these two GPUs?

A: The HD 7790 has 1024 MB of dedicated GDDR5 memory on a 128-bit bus, delivering 96.00 GB/s of bandwidth. The 680M uses system shared memory, with its bus width and bandwidth described as system dependent.

Q: Which GPU has the higher pixel and texture rates?

A: The Radeon 680M dominates both rates: 70.40 GPixel/s versus 16.00 GPixel/s for the HD 7790, and 105.6 GTexel/s versus 56.00 GTexel/s.

Q: What are the power connector requirements?

A: The HD 7790 requires a single 6-pin power connector and has a TDP of 85 W. The 680M is an integrated graphics processor (IGP) with no power connectors and a 50 W TDP.

Architecture Differences

The architectural divide between these two AMD parts is substantial. The HD 7790 is built on the Bonaire chip using GCN 2.0 architecture, part of the Southern Islands generation (HD 7700 series). It uses a 28 nm process from TSMC with 2,080 million transistors on a 160 mm² die, yielding a transistor density of 13.0 million per square millimeter. In contrast, the Radeon 680M uses the Rembrandt+ chip with RDNA 2.0 architecture, part of the Navi II IGP generation (Rembrandt Mobile). It is fabricated on a much newer 6 nm TSMC process and packs 13,100 million transistors on a 208 mm² die, achieving a density of 63.0 million per square millimeter.

The shader configuration differs as well. The HD 7790 has 896 shading units, 56 texture mapping units, and 16 raster operation units. The 680M has 768 shading units, 48 TMUs, and 32 ROPs. Notably, the 680M includes 12 ray tracing cores, a feature absent from the HD 7790 entirely.

Clock behavior separates them further. The HD 7790 lists no base or boost clock in the database, with memory running at 1500 MHz (6 Gbps effective). The 680M has a base clock of 2000 MHz and a boost clock of 2200 MHz, with system shared memory. The HD 7790 supports PCIe 3.0 x16, while the 680M uses PCIe 4.0 x8. API support also reflects their eras: the HD 7790 supports DirectX 12 (12_0) and Vulkan 1.2.170, whereas the 680M supports DirectX 12 Ultimate (12_2) and Vulkan 1.4.

The HD 7790 is a dual-slot, 183 mm (7.2 inches) discrete card with display outputs including DVI, HDMI 1.4a, and two mini-DisplayPort 1.2 connectors. The 680M is an IGP with no slot width and portable device dependent display outputs. The HD 7790 is end-of-life, released in March 2013, while the 680M remains active, released in January 2023.

The Verdict

The data points to a clear split: the HD 7790 wins on raw average benchmark score, while the 680M wins on modern feature set and efficiency. In the Geekbench Metal test, the HD 7790 scores 17,666, which places it just 0.2% above the RTX 4060 and 0.4% above the 780M. The 680M's average of 15,270 comes from a different benchmark mix: 3DMark Steel Nomad DX12 at 378, Geekbench OpenCL at 23,468, and Geekbench Vulkan at 21,965.

For users prioritizing discrete GPU performance in legacy workloads, the HD 7790 is the pick. Its single benchmark score is higher, and its percentile ranking (61 versus 57) confirms a stronger overall position among all GPUs. The 680M, however, offers ray tracing cores, a much higher pixel rate (70.40 GPixel/s versus 16.00 GPixel/s), more than double the FP32 throughput (3.379 TFLOPS versus 1.792 TFLOPS), and half the TDP (50 W versus 85 W). It also supports DirectX 12 Ultimate and Vulkan 1.4, making it the choice for modern API compatibility and integrated graphics scenarios. The HD 7790 is end-of-life; the 680M is active. For anyone building a new system today, the 680M's active status and feature advantages outweigh the HD 7790's higher average score.

Specification Differences

| Specification | AMD Radeon HD 7790 | AMD Radeon 680M |

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

| Chip | Bonaire | Rembrandt+ |

| Architecture | GCN 2.0 | RDNA 2.0 |

| Generation | Southern Islands (HD 7700) | Navi II IGP (Rembrandt Mobile) |

| Process Node | 28 nm | 6 nm |

| Transistors | 2,080 million | 13,100 million |

| Die Size | 160 mm² | 208 mm² |

| Transistor Density | 13.0M / mm² | 63.0M / mm² |

| Base Clock | None | 2000 MHz |

| Boost Clock | None | 2200 MHz |

| Memory | 1024 MB GDDR5 | System Shared |

| Memory Bus | 128 bit | System Shared |

| Memory Bandwidth | 96.00 GB/s | System Dependent |

| Shading Units | 896 | 768 |

| TMUs | 56 | 48 |

| ROPs | 16 | 32 |

| Ray Tracing Cores | None | 12 |

| Pixel Rate | 16.00 GPixel/s | 70.40 GPixel/s |

| Texture Rate | 56.00 GTexel/s | 105.6 GTexel/s |

| FP32 | 1.792 TFLOPS | 3.379 TFLOPS |

| FP16 | None | 6.758 TFLOPS (2:1) |

| TDP | 85 W | 50 W |

| Slot Width | Dual-slot | IGP |

| Power Connectors | 1x 6-pin | None |

| Suggested PSU | 250 W | None |

| Bus Interface | PCIe 3.0 x16 | PCIe 4.0 x8 |

| Display Outputs | 1x DVI, 1x HDMI 1.4a, 2x mini-DisplayPort 1.2 | Portable Device Dependent |

| DirectX | 12 (12_0) | 12 Ultimate (12_2) |

| Vulkan | 1.2.170 | 1.4 |

| Production Status | End-of-life | Active |

| Release Date | 2013-03-21 | 2023-01-02 |

| Launch MSRP | 149 USD | None |

| Predecessor | Northern Islands | Vega II IGP |

| Successor | Sea Islands | Navi III IGP |

Head-to-Head Benchmarks

The database records no direct head-to-head benchmark results between these two GPUs, so the comparison relies on their individual benchmark suites. The HD 7790's sole recorded benchmark is Geekbench Metal, scoring 17,666. That score places it 0.2% above the RTX 4060 and 0.7% above the Radeon Pro 560, while sitting 0.7% below the Quadro RTX 4000. The 680M's strongest recorded result is Geekbench OpenCL at 23,468, followed by Geekbench Vulkan at 21,965. Its 3DMark Steel Nomad DX12 score is much lower at 378, reflecting the different rendering workloads in that test.

The average scores tell the headline story. The HD 7790 averages 17,666, which is 2,396 points higher than the 680M's 15,270 average. That is a 15.7% advantage in raw aggregate performance. However, the 680M's individual OpenCL and Vulkan scores are substantially higher than the HD 7790's single Metal result. The 680M's OpenCL score is 5,802 points above the HD 7790's Metal score, and its Vulkan score is 4,299 points above.

The nearest rival comparisons reinforce the gap in positioning. The HD 7790's closest rival is the RTX 4060 at a 0.2% delta, meaning its performance is effectively interchangeable with a modern mid-range discrete GPU in the Metal test. The 680M's closest rival is the GTX 580 at a -0.1% delta, placing its average performance at the level of a much older flagship. Yet the 680M also edges out the RX 7600 by 0.7%, showing that its OpenCL and Vulkan strengths can overcome its weaker 3DMark result in certain APIs.

Where Each One Wins

The HD 7790 wins in aggregate benchmark performance. Its average score of 17,666 exceeds the 680M's 15,270, and its 61st percentile ranking beats the 680M's 57th. In the Geekbench Metal test specifically, the HD 7790's 17,666 score puts it ahead of the RTX 4060 and the 780M, two much newer GPUs. This makes it the stronger choice for applications that rely on Metal compute performance, where it performs on par with or slightly above those modern rivals.

The 680M wins in raw throughput metrics. Its FP32 of 3.379 TFLOPS is nearly double the HD 7790's 1.792 TFLOPS. Its FP16 capability of 6.758 TFLOPS is entirely absent on the HD 7790. The pixel rate of 70.40 GPixel/s is 4.4 times higher than the HD 7790's 16.00 GPixel/s, and the texture rate of 105.6 GTexel/s is 1.9 times higher. The 680M also has twice the ROPs (32 versus 16) and includes 12 ray tracing cores.

The 680M wins on efficiency and integration. Its 50 W TDP is 35 W lower than the HD 7790's 85 W. It requires no power connectors and no suggested PSU, functioning as an IGP. The HD 7790 needs a 6-pin connector and a 250 W suggested power supply. The 680M also uses a 6 nm process versus 28 nm, giving it a transistor density of 63.0M per square millimeter versus 13.0M.

For use cases, the HD 7790 is preferable for discrete desktop builds from its era, particularly where dedicated GDDR5 memory and a fixed 96.00 GB/s bandwidth matter. The 680M is preferable for modern portable devices, where its integrated nature, ray tracing support, DirectX 12 Ultimate, and Vulkan 1.4 compatibility align with current software requirements. The HD 7790's end-of-life status and 2013 release date make it a legacy option, while the 680M's active status and 2023 release date make it the forward-looking choice.