AMD Radeon RX 5700 XT vs NVIDIA GeForce GTX 780 Comparison

Where Each One Wins

The benchmark data splits cleanly between the two cards, with each taking a distinct set of workloads. The NVIDIA GeForce GTX 780 wins the Geekbench OpenCL test with a score of 22,863 against the AMD Radeon RX 5700 XT's 9,524, a massive 140.1% advantage. That is the only test where the older card comes out ahead. The AMD Radeon RX 5700 XT dominates the other two recorded comparisons, taking the Geekbench Metal test with 74,647 against 10,114, an 86.5% lead, and the Geekbench Vulkan test with 68,310 against 24,514, a 64.1% lead. So the use-case split is straightforward: for OpenCL compute workloads the GTX 780 is the stronger part, while for Metal and Vulkan graphics and compute paths the RX 5700 XT is decisively faster.

Looking at the broader database picture, the GTX 780 sits in the 64th percentile of all GPUs, with an average benchmark score of 19,164. Its nearest rivals include the NVIDIA TITAN Xp at 19,177 (0.1% higher), the NVIDIA Tesla K20m at 19,089 (0.4% lower), the NVIDIA GeForce RTX 4050 Mobile at 19,049 (0.6% lower), and the AMD Radeon RX 6600 at 19,036 (0.7% lower). The RX 5700 XT, by contrast, sits in the 59th percentile with an average score of 16,361. Its nearest rivals include the AMD Radeon Pro 5600M at 16,351 (0.1% higher), the NVIDIA RTX PRO 6000 Blackwell at 16,408 (0.3% higher), the AMD Radeon PRO W7500 at 16,415 (0.3% higher), and the NVIDIA GeForce RTX 5090 D V2 at 16,504 (0.9% higher). The interesting wrinkle is that the GTX 780 has the higher average score and percentile, yet the RX 5700 XT wins two of the three head-to-head tests by very large margins. This suggests the GTX 780's lead in the aggregate is carried almost entirely by its OpenCL performance, while the RX 5700 XT excels in the newer API workloads.

Architecture Differences

The two cards come from very different design eras and manufacturing approaches. The NVIDIA GeForce GTX 780 uses the GK110 chip built on Kepler architecture, produced on TSMC's 28 nm process. The chip packs 7,080 million transistors into a 561 mm² die, giving a transistor density of 12.6 million per square millimeter. The AMD Radeon RX 5700 XT uses the Navi 10 chip built on RDNA 1.0 architecture, produced on TSMC's 7 nm process. It contains 10,300 million transistors on a much smaller 251 mm² die, yielding a transistor density of 41.0 million per square millimeter. That density gap of more than three times is one of the clearest signs of how much manufacturing technology advanced between the two releases.

The memory subsystems differ substantially. The GTX 780 has 3 GB of GDDR5 on a 384-bit bus, with memory clocked at 1502 MHz and 6 Gbps effective, delivering 288.4 GB/s of bandwidth. The RX 5700 XT has 8 GB of GDDR6 on a 256-bit bus, with memory clocked at 1750 MHz and 14 Gbps effective, delivering 448.0 GB/s of bandwidth. Despite the narrower bus, the RX 5700 XT achieves significantly higher bandwidth thanks to the newer memory type and higher effective clock.

Compute resources also differ in configuration. The GTX 780 has 2,304 shading units, 192 texture mapping units, and 48 ROPs. The RX 5700 XT has 2,560 shading units, 160 TMUs, and 64 ROPs. The RX 5700 XT has more shaders and more ROPs, but fewer TMUs. The clock speeds are far apart: the GTX 780 runs at a base of 863 MHz and boost of 902 MHz, while the RX 5700 XT runs at a base of 1605 MHz, a game clock of 1755 MHz, and a boost of 1905 MHz. The higher clocks contribute directly to the RX 5700 XT's large leads in pixel rate, texture rate, and FP32 throughput.

Neither card has ray tracing cores or tensor cores. The GTX 780 supports DirectX 12 (11_0), OpenGL 4.6, and Vulkan 1.2.175. The RX 5700 XT supports DirectX 12 (12_1), OpenGL 4.6, and Vulkan 1.4. The RX 5700 XT also lists FP16 performance at 19.51 TFLOPS with a 2:1 ratio, while the GTX 780 has no recorded FP16 figure. The RX 5700 XT uses a PCIe 4.0 x16 interface, while the GTX 780 uses PCIe 3.0 x16. Display outputs also differ: the GTX 780 has 2x DVI, 1x HDMI 1.4a, and 1x DisplayPort 1.2, while the RX 5700 XT has 1x HDMI 2.0b and 3x DisplayPort 1.4a.

Head-to-Head Benchmarks

The recorded head-to-head tests cover three Geekbench workloads, and each tells a different story about the two architectures.

In Geekbench Metal, the AMD Radeon RX 5700 XT scores 74,647 versus the NVIDIA GeForce GTX 780's 10,114. That is an 86.5% difference in the AMD card's favor. Metal is Apple's graphics API, and the RX 5700 XT's RDNA 1.0 architecture clearly handles it far better than Kepler. The GTX 780's Metal score is so low relative to its other results that it likely reflects limited driver optimization for that API on the older architecture. This is not a close contest by any measure.

In Geekbench OpenCL, the tables turn completely. The NVIDIA GeForce GTX 780 scores 22,863 while the AMD Radeon RX 5700 XT scores only 9,524. That gives the GTX 780 a 140.1% advantage. OpenCL is an older compute standard, and the GTX 780's GK110 chip, which was originally designed with heavy compute workloads in mind, performs very strongly here. The RX 5700 XT's relatively low OpenCL score, despite its much higher raw FP32 throughput, suggests that its OpenCL driver path is not as well optimized as its Vulkan and Metal paths. The 140.1% delta is the largest margin in either direction across all three tests.

In Geekbench Vulkan, the AMD Radeon RX 5700 XT wins again, scoring 68,310 against the NVIDIA GeForce GTX 780's 24,514. That is a 64.1% lead. Vulkan is a modern low-level API, and the RX 5700 XT's RDNA architecture was designed for it. The GTX 780's Vulkan support, listed as version 1.2.175, is present but the hardware and driver stack are much older. The result is a decisive but slightly smaller margin than the Metal test.

Overall, the head-to-head record stands at 1 win for the GTX 780 and 2 wins for the RX 5700 XT. But the margins matter more than the count: the GTX 780's OpenCL win is enormous, while the RX 5700 XT's wins in Metal and Vulkan are also very large. Neither card wins by a small amount anywhere. The data points to two very different performance profiles rather than a near tie.

The Verdict

The recorded data shows that the NVIDIA GeForce GTX 780 and AMD Radeon RX 5700 XT are not direct competitors in the traditional sense, but rather two cards with opposite strengths. The GTX 780 is the clear choice for OpenCL compute workloads, where it beats the RX 5700 XT by 140.1%. It also holds a higher average benchmark score of 19,164 against the RX 5700 XT's 16,361, and a higher percentile ranking at 64 versus 59. For anyone running OpenCL-based applications, the GTX 780 is the better performer in this pairing.

The RX 5700 XT is the better card for modern graphics APIs. It wins Metal by 86.5% and Vulkan by 64.1%, which are the workloads most relevant to current gaming and GPU-accelerated applications. Its memory configuration also provides more headroom: 8 GB of GDDR6 with 448.0 GB/s of bandwidth versus 3 GB of GDDR5 with 288.4 GB/s. Its pixel rate of 121.9 GPixel/s is nearly three times the GTX 780's 43.30 GPixel/s, and its texture rate of 304.8 GTexel/s is well ahead of the GTX 780's 173.2 GTexel/s. FP32 throughput is also much higher at 9.754 TFLOPS versus 4.156 TFLOPS.

From the data, the choice depends entirely on the target API. OpenCL users should favor the GTX 780. Metal and Vulkan users should favor the RX 5700 XT. The RX 5700 XT also supports newer standards including DirectX 12_1, Vulkan 1.4, and PCIe 4.0, which gives it a longer useful life in modern systems. The GTX 780, despite its strong OpenCL showing, is limited to DirectX 11_0 and PCIe 3.0, which restricts its relevance in current software environments. The verdict is not about which card is generally faster, but about which API each card serves best.

FAQ

Q: Which card has the higher average benchmark score?

A: The NVIDIA GeForce GTX 780 has an average benchmark score of 19,164, while the AMD Radeon RX 5700 XT has an average of 16,361.

Q: How large is the GTX 780's OpenCL lead over the RX 5700 XT?

A: The GTX 780 scores 22,863 in Geekbench OpenCL compared to the RX 5700 XT's 9,524, a 140.1% advantage.

Q: What are the memory capacities and types of each card?

A: The GTX 780 has 3 GB of GDDR5 with a 384-bit bus and 288.4 GB/s bandwidth. The RX 5700 XT has 8 GB of GDDR6 with a 256-bit bus and 448.0 GB/s bandwidth.

Q: Which card has a higher pixel rate?

A: The RX 5700 XT has a pixel rate of 121.9 GPixel/s, compared to the GTX 780's 43.30 GPixel/s.

Q: What DirectX versions do the two cards support?

A: The GTX 780 supports DirectX 12 (11_0), while the RX 5700 XT supports DirectX 12 (12_1).

Q: How do the two cards compare in Vulkan performance?

A: The RX 5700 XT scores 68,310 in Geekbench Vulkan against the GTX 780's 24,514, a 64.1% lead.

Specification Differences

| Field | NVIDIA GeForce GTX 780 | AMD Radeon RX 5700 XT |

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

| Chip | GK110 | Navi 10 |

| Architecture | Kepler | RDNA 1.0 |

| Process node | 28 nm | 7 nm |

| Transistors | 7,080 million | 10,300 million |

| Die size | 561 mm² | 251 mm² |

| Transistor density | 12.6M / mm² | 41.0M / mm² |

| Base clock | 863 MHz | 1605 MHz |

| Boost clock | 902 MHz | 1905 MHz |

| Game clock | None | 1755 MHz |

| Memory clock | 1502 MHz, 6 Gbps effective | 1750 MHz, 14 Gbps effective |

| Memory size | 3 GB | 8 GB |

| Memory type | GDDR5 | GDDR6 |

| Memory bus width | 384 bit | 256 bit |

| Memory bandwidth | 288.4 GB/s | 448.0 GB/s |

| Shading units | 2304 | 2560 |

| TMUs | 192 | 160 |

| ROPs | 48 | 64 |

| Pixel rate | 43.30 GPixel/s | 121.9 GPixel/s |

| Texture rate | 173.2 GTexel/s | 304.8 GTexel/s |

| FP32 | 4.156 TFLOPS | 9.754 TFLOPS |

| FP16 | None | 19.51 TFLOPS (2:1) |

| TDP | 250 W | 225 W |

| Suggested PSU | 600 W | 550 W |

| Bus interface | PCIe 3.0 x16 | PCIe 4.0 x16 |

| Display outputs | 2x DVI, 1x HDMI 1.4a, 1x DisplayPort 1.2 | 1x HDMI 2.0b, 3x DisplayPort 1.4a |

| DirectX | 12 (11_0) | 12 (12_1) |

| Vulkan | 1.2.175 | 1.4 |

| Length | 267 mm (10.5 inches) | 272 mm (10.7 inches) |

| Height | 111 mm (4.4 inches) | 111 mm (4.4 inches) |

| Width | 38 mm (1.5 inches) | 36 mm (1.4 inches) |

| Release date | 2013-05-22 | 2019-07-06 |

| Launch MSRP | 649 USD | 399 USD |