NVIDIA GeForce GTX 780 Ti vs NVIDIA GeForce RTX 2080 SUPER Comparison

The NVIDIA GeForce GTX 780 Ti and the NVIDIA GeForce RTX 2080 SUPER are separated by nearly six years of architecture evolution, yet their aggregate benchmark scores place them in the same performance tier. The GTX 780 Ti posts an average benchmark score of 24236, while the RTX 2080 SUPER scores 24170, a difference of only 0.3% in favor of the older card. However, this statistical parity masks a fundamental divergence in workload capabilities. The RTX 2080 SUPER dominates modern compute and graphics APIs, while the GTX 780 Ti’s edge in aggregate score comes from legacy benchmark results. The data indicates that the RTX 2080 SUPER is the superior choice for contemporary software, but the GTX 780 Ti remains competitive in specific legacy scenarios.

The Verdict

Based strictly on the benchmark data, the RTX 2080 SUPER is the clear winner for users running modern applications. In the two head-to-head tests available, the RTX 2080 SUPER outperforms the GTX 780 Ti by 72.5% in Geekbench OpenCL (99226 vs 27326) and by 75.5% in Geekbench Vulkan (111284 vs 27238). These are not marginal victories; they represent a more than threefold increase in compute throughput. The GTX 780 Ti’s only claim to superiority is its slightly higher average benchmark score of 24236, which is 0.3% ahead of the RTX 2080 SUPER’s 24170. This aggregate figure is misleading, as it is derived from a smaller set of older tests that do not exercise the newer card’s full feature set.

For users focused on DirectX 12 Ultimate, Vulkan 1.4, or ray tracing, the RTX 2080 SUPER is the only viable option. The GTX 780 Ti supports DirectX 12 (11_1) and Vulkan 1.2.175, but lacks the hardware features that modern games increasingly require. The RTX 2080 SUPER also offers 8 GB of GDDR6 memory versus 3 GB of GDDR5, which matters for texture-heavy workloads at high resolutions. If the workload is limited to older DirectX 11 titles or OpenGL applications, the GTX 780 Ti can still hold its own, as evidenced by its comparable aggregate score. However, the lack of modern API support and the massive deficit in compute benchmarks make the RTX 2080 SUPER the logical recommendation for any new purchase or upgrade.

FAQ

Q: Which card has a higher average benchmark score?

A: The NVIDIA GeForce GTX 780 Ti has an average benchmark score of 24236, which is 0.3% higher than the NVIDIA GeForce RTX 2080 SUPER’s score of 24170.

Q: How large is the performance gap in Vulkan compute?

A: In the Geekbench Vulkan test, the RTX 2080 SUPER scores 111284, while the GTX 780 Ti scores 27238. The RTX 2080 SUPER leads by 75.5%.

Q: Do both cards support the same DirectX version?

A: No. The RTX 2080 SUPER supports DirectX 12 Ultimate (12_2), while the GTX 780 Ti supports DirectX 12 (11_1).

Q: What is the memory capacity difference?

A: The RTX 2080 SUPER comes with 8 GB of GDDR6 memory, while the GTX 780 Ti has 3 GB of GDDR5 memory.

Q: Which card has a higher pixel fill rate?

A: The RTX 2080 SUPER has a pixel rate of 116.2 GPixel/s, which is more than double the GTX 780 Ti’s 55.68 GPixel/s.

Q: Are both cards considered end-of-life products?

A: Yes, both the GTX 780 Ti and the RTX 2080 SUPER have a production status of "End-of-life."

Architecture Differences

The GTX 780 Ti is built on the Kepler architecture using the GK110B chip, fabricated on a 28 nm process at TSMC. It contains 7,080 million transistors on a 561 mm² die, yielding a transistor density of 12.6M per mm². The RTX 2080 SUPER uses the Turing architecture with the TU104 chip, also from TSMC but on a 12 nm process. This newer chip packs 13,600 million transistors into a slightly smaller 545 mm² die, achieving a transistor density of 25.0M per mm². The density nearly doubles, reflecting the more advanced manufacturing node.

The most significant architectural divergence is the inclusion of dedicated hardware in the RTX 2080 SUPER. It features 48 RT cores and 384 tensor cores, which enable hardware-accelerated ray tracing and AI-based operations. The GTX 780 Ti has no such units, relying entirely on its 2880 shading units for all processing. The RTX 2080 SUPER also has more shading units (3072) and ROPs (64 vs 48), but fewer TMUs (192 vs 240). The memory subsystems are fundamentally different: the GTX 780 Ti uses a 384-bit GDDR5 interface, while the RTX 2080 SUPER uses a 256-bit GDDR6 interface. Despite the narrower bus, the newer memory technology delivers significantly higher bandwidth.

The API support highlights the generational gap. The GTX 780 Ti supports DirectX 12 (11_1), OpenGL 4.6, and Vulkan 1.2.175. The RTX 2080 SUPER supports DirectX 12 Ultimate (12_2), OpenGL 4.6, and Vulkan 1.4. The Turing card also introduces a 2:1 FP16 ratio, offering 22.30 TFLOPS of half-precision compute, a feature absent from the Kepler card’s specifications.

Specification Differences

The two cards share identical power requirements: both have a TDP of 250 W, use a 1x 6-pin + 1x 8-pin power connector setup, and recommend a 600 W power supply. They also share the same physical length of 267 mm and dual-slot width. The dimensions differ slightly in height and width: the GTX 780 Ti is 111 mm tall and 38 mm wide, while the RTX 2080 SUPER is 116 mm tall and 35 mm wide.

Core clock speeds are substantially different. The GTX 780 Ti runs at a base of 875 MHz with a boost of 928 MHz, while the RTX 2080 SUPER operates at a base of 1650 MHz and a boost of 1815 MHz. Memory clocks also diverge: the GTX 780 Ti uses 1753 MHz (7 Gbps effective), while the RTX 2080 SUPER uses 1937 MHz (15.5 Gbps effective). The RTX 2080 SUPER’s memory bandwidth is 495.9 GB/s, compared to 336.6 GB/s for the GTX 780 Ti.

Display outputs differ as well. The GTX 780 Ti offers 2x DVI, 1x HDMI 1.4a, and 1x DisplayPort 1.2. The RTX 2080 SUPER provides 1x HDMI 2.0, 3x DisplayPort 1.4a, and 1x USB Type-C. The RTX 2080 SUPER also has a higher compute throughput: 11.15 TFLOPS FP32 versus 5.345 TFLOPS, and a texture rate of 348.5 GTexel/s versus 222.7 GTexel/s. Both cards were released with a launch MSRP of 699 USD.

Head-to-Head Benchmarks

The head-to-head data includes only two tests, both of which the RTX 2080 SUPER wins decisively. In Geekbench OpenCL, the RTX 2080 SUPER scores 99226, while the GTX 780 Ti scores 27326. The deltaPct is -72.5%, meaning the GTX 780 Ti trails by that margin. In Geekbench Vulkan, the gap widens slightly: the RTX 2080 SUPER scores 111284 against the GTX 780 Ti’s 27238, with a deltaPct of -75.5%.

These results underscore the architectural advantage of Turing in compute-heavy workloads. The OpenCL score of 99226 is more than 3.6 times higher than the GTX 780 Ti’s score, and the Vulkan score is more than 4 times higher. Even accounting for the GTX 780 Ti’s higher average benchmark score, which is driven by its Geekbench Metal result of 18144 (a test the RTX 2080 SUPER does not have), the compute gap is insurmountable for modern applications. The RTX 2080 SUPER’s Passmark scores further illustrate its strength: 19490 in G3D, 8290 in GPU compute, and 920 in G2D. The GTX 780 Ti has no Passmark data in the fact pack, so direct comparison is limited to the two Geekbench tests.

The wins are split 2-0 in favor of the RTX 2080 SUPER. The GTX 780 Ti’s percentile rank of 70 versus the RTX 2080 SUPER’s 69 indicates that the older card sits slightly higher in the overall database, but this is a statistical artifact of the available test suite rather than a reflection of real-world capability.

Where Each One Wins

The RTX 2080 SUPER wins in every scenario that leverages modern APIs or parallel compute. Its 75.5% lead in Vulkan and 72.5% lead in OpenCL make it the definitive choice for GPU-accelerated workloads such as machine learning inference, video encoding, or scientific simulation. The presence of 48 RT cores and 384 tensor cores means it can handle ray-traced rendering and DLSS-style upscaling, features the GTX 780 Ti cannot execute at all. The 8 GB memory capacity and 495.9 GB/s bandwidth also give it a significant advantage in high-resolution texture loading and multi-tasking.

The GTX 780 Ti’s wins are confined to the aggregate benchmark score, where it edges out the RTX 2080 SUPER by 0.3%. This is driven by its Geekbench Metal score of 18144, which is not present in the RTX 2080 SUPER’s benchmark list. In legacy DirectX 9, 10, or 11 applications, the GTX 780 Ti may perform adequately, as evidenced by the RTX 2080 SUPER’s relatively low Passmark DirectX 9 score of 227 and DirectX 10 score of 140. However, these are the RTX 2080 SUPER’s numbers, not the GTX 780 Ti’s, so direct comparison is impossible from the data.

For users with older software that relies on Kepler-era compute or specific OpenGL features, the GTX 780 Ti remains a functional option. Its 3 GB of memory and 384-bit bus provide solid bandwidth for 1080p gaming in older titles. But for any forward-looking use case, the RTX 2080 SUPER is the only choice. The data is unambiguous: in the tests that matter for current and future software, the RTX 2080 SUPER is between 72.5% and 75.5% faster. The GTX 780 Ti’s aggregate score is a historical curiosity, not a competitive advantage.