NVIDIA GeForce GTX 780 Ti vs NVIDIA GeForce RTX 3070 Ti Mobile Comparison

The GTX 780 Ti and RTX 3070 Ti Mobile represent two very different eras of NVIDIA graphics, separated by nearly a decade of architectural evolution. The data shows a clear generational gap: in the two shared benchmark tests, the RTX 3070 Ti Mobile wins decisively, delivering roughly 3.9x the OpenCL score and 3.7x the Vulkan score of the older card. The GTX 780 Ti, however, still holds its own as a desktop part with a higher average benchmark score than its mobile rival, though this is likely due to the differing benchmark suites applied to each. The RTX 3070 Ti Mobile sits at the 69th percentile of all GPUs, while the GTX 780 Ti sits at the 70th, indicating that despite the massive performance delta in raw compute, the older card remains contextually competitive within its own benchmark set.

Head-to-Head Benchmarks

The single most dramatic result in the shared data is the Geekbench OpenCL test. The RTX 3070 Ti Mobile scores 106,022 points, while the GTX 780 Ti scores 27,326, a delta of -74.2% from the perspective of the older card. This means the RTX 3070 Ti Mobile is approximately 288% faster in raw compute throughput. This is not a marginal improvement; it is a generational leap that reflects the fundamental shift in architecture, memory bandwidth, and core count between the two.

In the Geekbench Vulkan test, the pattern repeats. The RTX 3070 Ti Mobile achieves 100,027 points against the GTX 780 Ti’s 27,238, a delta of -72.8%. Vulkan is a low-level API that benefits from modern hardware features and driver optimization, and the Ampere architecture’s support for Vulkan 1.4 (versus Kepler’s 1.2.175) clearly pays off. The RTX 3070 Ti Mobile’s advantage here is slightly smaller than in OpenCL but still overwhelming, suggesting that the raw compute advantage is consistent across different API workloads.

The GTX 780 Ti’s best showing in the head-to-head is simply that it shows up; its wins count is 0, while the RTX 3070 Ti Mobile wins 2. There is no benchmark in the shared set where the older card comes out ahead. However, the GTX 780 Ti’s average benchmark score across its own suite (Geekbench Metal, OpenCL, and Vulkan) is 24,236, which is higher than the RTX 3070 Ti Mobile’s average of 23,518 across its much broader suite (including 3DMark, Passmark, and Geekbench). This discrepancy is a reminder that different benchmark methodologies can produce different aggregate pictures, and the mobile part’s lower average is influenced by its inclusion of more diverse tests like Passmark DirectX 9 and G2D, which are not comparable to the desktop card’s tests.

FAQ

Q: Which GPU has the higher average benchmark score?

A: The NVIDIA GeForce GTX 780 Ti has a higher average benchmark score of 24,236, compared to the RTX 3070 Ti Mobile’s 23,518. This is despite the RTX 3070 Ti Mobile winning every shared head-to-head test by a wide margin.

Q: What is the performance difference in Vulkan compute workloads?

A: The RTX 3070 Ti Mobile scores 100,027 in Geekbench Vulkan, while the GTX 780 Ti scores 27,238, making the newer card approximately 267% faster in this specific test, based on the -72.8% delta.

Q: How do the two cards compare in terms of memory bandwidth?

A: The RTX 3070 Ti Mobile offers 448.0 GB/s of bandwidth from its 8 GB GDDR6 memory on a 256-bit bus, while the GTX 780 Ti provides 336.6 GB/s from 3 GB GDDR5 on a 384-bit bus. The mobile part’s bandwidth is roughly 33% higher.

Q: Which card has more shading units?

A: The RTX 3070 Ti Mobile has 5,888 shading units, more than double the GTX 780 Ti’s 2,880. This directly contributes to its FP32 compute advantage of 16.60 TFLOPS versus 5.345 TFLOPS.

Q: What are the TDP differences?

A: The GTX 780 Ti has a TDP of 250 W, while the RTX 3070 Ti Mobile is rated at 115 W. This means the mobile card delivers vastly higher performance while consuming less than half the power, a key benefit of the newer 8 nm process node.

Q: Do both cards support DirectX 12?

A: Yes, but at different feature levels. The GTX 780 Ti supports DirectX 12 (11_1), while the RTX 3070 Ti Mobile supports DirectX 12 Ultimate (12_2), which includes hardware ray tracing and mesh shaders.

Architecture Differences

The architectural gap between these two GPUs is vast. The GTX 780 Ti is built on the Kepler architecture, using the GK110B chip fabricated on a 28 nm process at TSMC. It packs 7,080 million transistors into a 561 mm² die, yielding a transistor density of 12.6 million transistors per square millimeter. In contrast, the RTX 3070 Ti Mobile uses the Ampere architecture with the GA104 chip, built on Samsung’s 8 nm process. It contains 17,400 million transistors on a smaller 392 mm² die, achieving a density of 44.4 million transistors per square millimeter—over 3.5x the density of Kepler.

This density difference explains everything else. The RTX 3070 Ti Mobile features 5,888 shading units, 184 texture mapping units, and 96 ROPs, versus 2,880 shading units, 240 TMUs, and 48 ROPs on the GTX 780 Ti. The newer card also introduces dedicated hardware that the older card lacks entirely: 46 ray tracing cores and 184 tensor cores. These are essential for modern features like DLSS and real-time ray tracing, which the GTX 780 Ti cannot accelerate.

Memory technology also diverges sharply. The GTX 780 Ti uses 3 GB of GDDR5 at 7 Gbps effective, while the RTX 3070 Ti Mobile uses 8 GB of GDDR6 at 14 Gbps effective. Despite the older card’s wider 384-bit bus, the newer card’s higher-speed memory on a 256-bit bus yields 448.0 GB/s versus 336.6 GB/s. The FP32 compute figures are even more lopsided: 16.60 TFLOPS for Ampere versus 5.345 TFLOPS for Kepler. The Ampere card also offers FP16 at 16.60 TFLOPS (1:1), while the Kepler card has no listed FP16 capability.

The API support tells a similar story of progression. The GTX 780 Ti supports DirectX 12 (11_1), OpenGL 4.6, and Vulkan 1.2.175. The RTX 3070 Ti Mobile supports DirectX 12 Ultimate (12_2), OpenGL 4.6, and Vulkan 1.4. The newer card’s PCIe interface is also faster: PCIe 4.0 x16 versus PCIe 3.0 x16. The GTX 780 Ti requires substantial power delivery with 1x 6-pin and 1x 8-pin connectors, a 600 W suggested PSU, and a dual-slot cooler, while the RTX 3070 Ti Mobile has no power connectors listed, consistent with its mobile design.

The Verdict

The data is unambiguous: the RTX 3070 Ti Mobile is the superior performer in every shared benchmark, and its architectural advantages in memory, compute, and features are overwhelming. The GTX 780 Ti, however, still holds a higher average benchmark score across its own test suite, which suggests that in legacy or specific workloads (like Geekbench Metal, where it scores 18,144), it remains a capable part. The RTX 3070 Ti Mobile’s nearest rivals include the RTX 3080 Mobile (deltaPct -0.5) and the GTX TITAN Z (deltaPct -0.9), placing it in a high-end mobile tier. The GTX 780 Ti’s rivals include the GTX 1630 (deltaPct -0.2) and RTX 2080 SUPER (deltaPct 0.3), showing it clusters with mid-range desktop parts from later generations.

Users should pick the RTX 3070 Ti Mobile for any modern workload that benefits from raw compute, high memory bandwidth, or feature support like ray tracing. The GTX 780 Ti should only be considered if the workload is specifically optimized for its older architecture or if the user is constrained to its desktop form factor and legacy PCIe 3.0 interface. The GTX 780 Ti’s launch MSRP was 699 USD, but its end-of-life status and lack of modern features make it a hard sell against the mobile Ampere part.

Specification Differences

| Specification | GTX 780 Ti | RTX 3070 Ti Mobile |

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

| Architecture | Kepler | Ampere |

| Process Node | 28 nm | 8 nm |

| Transistors | 7,080 million | 17,400 million |

| Die Size | 561 mm² | 392 mm² |

| Base Clock | 875 MHz | 915 MHz |

| Boost Clock | 928 MHz | 1410 MHz |

| Memory Size | 3 GB | 8 GB |

| Memory Type | GDDR5 | GDDR6 |

| Memory Bus | 384 bit | 256 bit |

| Memory Bandwidth | 336.6 GB/s | 448.0 GB/s |

| Shading Units | 2880 | 5888 |

| TMUs | 240 | 184 |

| ROPs | 48 | 96 |

| RT Cores | None | 46 |

| Tensor Cores | None | 184 |

| FP32 Performance | 5.345 TFLOPS | 16.60 TFLOPS |

| TDP | 250 W | 115 W |

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

| Vulkan Support | 1.2.175 | 1.4 |

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

| Power Connectors | 1x 6-pin + 1x 8-pin | None |

Where Each One Wins

The RTX 3070 Ti Mobile wins in every measurable dimension of performance in the shared benchmarks. Its OpenCL score of 106,022 is nearly four times the GTX 780 Ti’s 27,326, making it the clear choice for compute-heavy applications like rendering, machine learning inference, or scientific simulation that leverage OpenCL. Its Vulkan performance is similarly dominant, with 100,027 points versus 27,238, which matters for modern games and applications that use Vulkan as their primary API. The newer card also wins on memory capacity (8 GB vs 3 GB) and bandwidth (448.0 GB/s vs 336.6 GB/s), which is critical for high-resolution textures and large datasets.

The GTX 780 Ti wins only in the narrow sense of its higher average benchmark score (24,236 vs 23,518), driven by its strong Geekbench Metal result of 18,144. This suggests that in Apple’s Metal API workloads, the older desktop card may still be competitive—though there is no direct head-to-head Metal benchmark in the data. The GTX 780 Ti also has a higher texture rate in terms of TMU count (240 vs 184), though the newer card compensates with higher clock speeds and a higher overall texture rate (259.4 GTexel/s vs 222.7 GTexel/s). The GTX 780 Ti’s fixed desktop form factor with dual-slot cooling and standard display outputs (2x DVI, HDMI 1.4a, DisplayPort 1.2) makes it easier to deploy in a traditional tower, whereas the RTX 3070 Ti Mobile is portable-device dependent. For anyone building a new system or upgrading for modern titles, the RTX 3070 Ti Mobile is the only rational choice from a pure performance standpoint.