NVIDIA GeForce GTX 770 vs NVIDIA GeForce RTX 5090 D V2 Comparison

Head-to-Head Benchmarks

The two cards in this comparison do not share a single common benchmark in the database, so the head-to-head picture must be assembled from their respective recorded averages and closest rivals. The RTX 5090 D V2 posts an average benchmark score of 16,504, while the GTX 770 averages 14,747. That is a raw gap of 1,757 points, or roughly 11.9% higher for the newer card. The data shows the RTX 5090 D V2 sits at the 59th percentile among all GPUs, while the GTX 770 sits at the 57th percentile. The percentile difference is modest, but the actual workload behavior diverges sharply when you look at what each card was tested on.

The RTX 5090 D V2 has a single recorded result in 3DMark Steel Nomad DX12, scoring 16,504. That score places it almost exactly level with the NVIDIA T400 (16,508, a 0% delta), slightly ahead of the AMD Radeon PRO W7500 (16,415, a 0.5% delta), and marginally ahead of the NVIDIA RTX PRO 6000 Blackwell (16,408, a 0.6% delta). It also edges out the AMD Radeon RX 5700 XT (16,361, a 0.9% delta). In practical terms, the RTX 5090 D V2 is not dramatically faster than these rivals in this specific DX12 test; it is within a single percentage point of all four. The takeaway is that this card's strength is not in a universal thrashing of everything nearby, but rather in a consistent, narrow advantage across a cluster of similarly scoring products.

The GTX 770, by contrast, has three recorded results: Geekbench Metal at 9,369, Geekbench OpenCL at 15,707, and Geekbench Vulkan at 19,166. Its average of 14,747 is pulled down by the Metal result, which is far lower than its Vulkan result. The GTX 770's nearest rivals show a tighter grouping: the AMD Radeon RX 5500 XT scores 14,692 (a 0.4% delta), the NVIDIA Quadro M2000 scores 14,532 (a 1.5% delta), while the NVIDIA GeForce GTX 660 Ti scores 15,063 (a -2.1% delta relative to the GTX 770) and the AMD Radeon Pro 560X scores 15,082 (a -2.2% delta). So the GTX 770 actually leads two of its four closest rivals, but trails the GTX 660 Ti and the Radeon Pro 560X by about 2%. The data suggests the GTX 770 is competitive with its immediate peers, but not dominant.

The biggest wins each way are not available as direct matchups because the database has no overlapping tests. However, the recorded numbers indicate that the RTX 5090 D V2 wins on overall average score by a clear margin, and it wins decisively on memory bandwidth, pixel rate, texture rate, and raw FP32 throughput. The GTX 770 wins on nothing in the recorded metrics except for a lower power draw and a smaller physical footprint, which are not performance wins. The RTX 5090 D V2's 104.8 TFLOPS of FP32 is more than 31 times the GTX 770's 3.333 TFLOPS. Its 1.34 TB/s memory bandwidth is nearly six times the GTX 770's 224.4 GB/s. In any compute-heavy or memory-bound scenario, the data shows a landslide, not a close race.

Architecture Differences

The RTX 5090 D V2 is built on the GB202 chip using the Blackwell 2.0 architecture, fabricated on a 5 nm process at TSMC. It packs 92,200 million transistors into a 750 mm² die, yielding a transistor density of 122.9 million per mm². The GTX 770 uses the GK104 chip with the Kepler architecture, on a 28 nm process, also at TSMC, with 3,540 million transistors on a 294 mm² die, for a density of 12.0 million per mm². The transistor count difference is roughly 26-to-1 in favor of the newer card, and the density difference is about 10-to-1. That alone explains why the RTX 5090 D V2 can field 21,760 shading units versus the GTX 770's 1,536, and 680 texture mapping units versus 128.

Memory architecture differs completely. The RTX 5090 D V2 uses 24 GB of GDDR7 on a 384-bit bus, with a bandwidth of 1.34 TB/s. The GTX 770 uses 2 GB of GDDR5 on a 256-bit bus, with 224.4 GB/s. The newer card also has 170 ray tracing cores and 680 tensor cores; the GTX 770 has none of either. The GTX 770 has 32 ROPs and a pixel rate of 34.72 GPixel/s, while the RTX 5090 D V2 has 176 ROPs and a pixel rate of 423.6 GPixel/s. Texture rate is 1,636.8 GTexel/s versus 138.9 GTexel/s. The FP32 compute is 104.8 TFLOPS (with FP16 at 1:1 ratio) versus 3.333 TFLOPS.

The RTX 5090 D V2 supports DirectX 12 Ultimate (12_2), OpenGL 4.6, and Vulkan 1.4. The GTX 770 supports DirectX 12 (11_0), OpenGL 4.6, and Vulkan 1.2.175. The newer card also uses a PCIe 5.0 x16 interface, while the GTX 770 uses PCIe 3.0 x16. Display outputs differ: the RTX 5090 D V2 has one HDMI 2.1b and three DisplayPort 2.1b outputs; the GTX 770 has two DVI, one HDMI 1.4a, and one DisplayPort 1.2. The power delivery is also distinct, with the RTX 5090 D V2 using a single 16-pin connector and the GTX 770 using one 6-pin plus one 8-pin.

FAQ

Q: Which card has a higher average benchmark score?

A: The RTX 5090 D V2 averages 16,504, while the GTX 770 averages 14,747. The newer card leads by 1,757 points, or about 11.9%.

Q: Does the GTX 770 have any ray tracing or tensor cores?

A: No. The GTX 770 has no ray tracing cores and no tensor cores. The RTX 5090 D V2 has 170 ray tracing cores and 680 tensor cores.

Q: How much memory does each card have, and what type?

A: The RTX 5090 D V2 has 24 GB of GDDR7 on a 384-bit bus. The GTX 770 has 2 GB of GDDR5 on a 256-bit bus.

Q: What is the FP32 compute difference?

A: The RTX 5090 D V2 delivers 104.8 TFLOPS of FP32, while the GTX 770 delivers 3.333 TFLOPS. That is a ratio of roughly 31.4 to 1.

Q: Which card is closer to its nearest rivals in performance?

A: The RTX 5090 D V2 is within 0.9% of its top four rivals, with deltas of 0%, 0.5%, 0.6%, and 0.9%. The GTX 770 has a wider spread, with deltas of 0.4%, 1.5%, -2.1%, and -2.2% against its closest rivals.

Q: What is the power draw for each card?

A: The RTX 5090 D V2 has a TDP of 575 W and a suggested PSU of 950 W. The GTX 770 has a TDP of 230 W and a suggested PSU of 550 W.

Specification Differences

The two cards differ in nearly every measurable specification. The RTX 5090 D V2 uses a 5 nm process; the GTX 770 uses 28 nm. Transistor count is 92,200 million versus 3,540 million. Die size is 750 mm² versus 294 mm². Transistor density is 122.9M per mm² versus 12.0M per mm². Base clock is 2017 MHz versus 1046 MHz; boost clock is 2407 MHz versus 1085 MHz. Memory clock is 1750 MHz (28 Gbps effective) versus 1753 MHz (7 Gbps effective). Memory size is 24 GB versus 2 GB; memory type is GDDR7 versus GDDR5; bus width is 384 bit versus 256 bit; bandwidth is 1.34 TB/s versus 224.4 GB/s.

Shading units are 21,760 versus 1,536. TMUs are 680 versus 128. ROPs are 176 versus 32. RT cores are 170 versus none. Tensor cores are 680 versus none. Pixel rate is 423.6 GPixel/s versus 34.72 GPixel/s. Texture rate is 1,636.8 GTexel/s versus 138.9 GTexel/s. FP32 is 104.8 TFLOPS versus 3.333 TFLOPS. FP16 is 104.8 TFLOPS (1:1) versus not applicable, as the GTX 770 has no recorded FP16 figure. TDP is 575 W versus 230 W. Power connectors are 1x 16-pin versus 1x 6-pin plus 1x 8-pin. Suggested PSU is 950 W versus 550 W. Bus interface is PCIe 5.0 x16 versus PCIe 3.0 x16. Display outputs are 1x HDMI 2.1b and 3x DisplayPort 2.1b versus 2x DVI, 1x HDMI 1.4a, and 1x DisplayPort 1.2. DirectX support is 12 Ultimate (12_2) versus 12 (11_0). Vulkan support is 1.4 versus 1.2.175. Dimensions are 304 mm length, 137 mm height, 48 mm width versus 267 mm length, 111 mm height, 38 mm width. Production status is Active versus End-of-life. Release date is 2025-08-14 versus 2013-05-29. The launch MSRP is 2,299 USD for the RTX 5090 D V2 and 399 USD for the GTX 770.

The Verdict

The data makes the choice unambiguous for modern workloads. The RTX 5090 D V2 is in a completely different performance class, with 31 times the FP32 compute, six times the memory bandwidth, and 12 times the pixel rate of the GTX 770. It also brings ray tracing, tensor cores, and modern API support that the GTX 770 simply lacks. The GTX 770 is end-of-life, while the RTX 5090 D V2 is active in production.

The GTX 770 does have one advantage: power consumption. At 230 W versus 575 W, it draws less than half the power of the newer card, and its suggested PSU is 550 W versus 950 W. For a builder with an old system that cannot handle a 575 W card, the GTX 770 might still run, but the performance gap is so enormous that it is hard to recommend the older card for anything beyond legacy compatibility. The GTX 770's nearest rival deltas show it is competitive with its own generation, but that generation is a decade old.

The RTX 5090 D V2's nearest rivals are all within 0.9% of its score, which indicates that in the specific DX12 test recorded, it is not head-and-shoulders above its immediate competitors. But those competitors are modern professional and consumer cards. The GTX 770's nearest rivals are older cards like the GTX 660 Ti and the Radeon Pro 560X, and it is within 2.2% of them. The data shows the RTX 5090 D V2 is the right choice for anyone running current games, compute workloads, or ray-traced content. The GTX 770 is only relevant for retro builds or systems with strict power limits.

Where Each One Wins

The RTX 5090 D V2 wins in every performance category measured: average benchmark score (16,504 versus 14,747), FP32 compute (104.8 TFLOPS versus 3.333 TFLOPS), texture rate (1,636.8 GTexel/s versus 138.9 GTexel/s), pixel rate (423.6 GPixel/s versus 34.72 GPixel/s), memory bandwidth (1.34 TB/s versus 224.4 GB/s), and memory capacity (24 GB versus 2 GB). It also wins on API support, with DirectX 12 Ultimate and Vulkan 1.4 versus the GTX 770's DirectX 12 (11_0) and Vulkan 1.2.175. It has ray tracing and tensor cores, which the GTX 770 does not have at all. It uses a newer PCIe 5.0 interface versus PCIe 3.0, and it has more display outputs with newer standards (HDMI 2.1b and DisplayPort 2.1b versus DVI, HDMI 1.4a, and DisplayPort 1.2).

The GTX 770 wins in exactly one meaningful area: power efficiency in absolute terms. It draws 230 W versus 575 W, and it only needs a 550 W PSU versus 950 W. It is also physically smaller, at 267 mm length versus 304 mm, 111 mm height versus 137 mm, and 38 mm width versus 48 mm. That makes it easier to fit in compact cases and older systems. It also uses standard 6-pin and 8-pin power connectors, which are more universally compatible than the newer 16-pin connector on the RTX 5090 D V2. For a builder with a 550 W power supply and a small case, the GTX 770 is the only one of the two that will physically work. But for anyone who can accommodate the power and size requirements, the RTX 5090 D V2 is the clear winner across every benchmark and specification that matters for performance.