NVIDIA GeForce GTX 690 vs NVIDIA GeForce GTX 770 Comparison

Head-to-Head Benchmarks

The two GeForce cards share a common silicon heritage, but their benchmark results tell a story of divergent tuning priorities. In the database's recorded tests, the GTX 690 and GTX 770 split their two head-to-head matchups exactly one win apiece, with the margins in both directions being substantial.

Starting with Geekbench OpenCL, the GTX 690 posts a score of 17,399 against the GTX 770's 15,707. That is a 10.8% advantage for the older, dual-GPU flagship. The GTX 690's OpenCL result is not just a narrow edge; it is a decisive gap that places it ahead of several modern cards in the overall database. Its average benchmark score of 17,037 puts it in the 60th percentile of all GPUs tracked. By comparison, the GTX 770's OpenCL score of 15,707 is its weakest recorded result across all tests, and its overall average of 14,747 lands at the 57th percentile. The OpenCL win for the GTX 690 aligns with its higher raw compute ceiling, as its FP32 throughput of 3.130 TFLOPS, while slightly lower than the GTX 770's 3.333 TFLOPS, is paired with dual-GPU scaling that clearly benefits compute-heavy workloads in this particular API.

The Vulkan test flips the script entirely. Here the GTX 770 scores 19,166, while the GTX 690 manages 16,675. That is a 13% swing in favor of the newer card, and notably, the GTX 770's Vulkan result is its best of all recorded benchmarks, exceeding its OpenCL score by more than 3,400 points. The GTX 690's Vulkan figure of 16,675 is actually lower than its OpenCL score, suggesting that the dual-GPU configuration does not scale as effectively under Vulkan's explicit multi-adapter model, or that driver maturity at the time of testing favored the single-GPU GTX 770. The delta of 13% is the largest margin in either direction between these two cards, making the Vulkan test the clearest differentiator in the head-to-head data.

Looking at the broader competitive context, the GTX 690's average score of 17,037 places it within a tight cluster of modern and professional hardware. It sits just 0.1% above the AMD Radeon HD 7970M (17,019) and 0.6% above the NVIDIA Tesla M4 (16,932), while trailing the AMD Radeon RX 7600 XT by a mere 0.3% (17,083) and the NVIDIA GeForce RTX 3070 by 1% (17,208). This is remarkable for a card from the GeForce 600 generation: it is essentially trading blows with an RTX 3070 in aggregate compute benchmarks. The GTX 770, on the other hand, anchors its average of 14,747 against a different tier of rivals. It leads the AMD Radeon RX 5500 XT by 0.4% (14,692) and the NVIDIA Quadro M2000 by 1.5% (14,532), while sitting 2.1% behind the GeForce GTX 660 Ti (15,063) and 2.2% behind the AMD Radeon Pro 560X (15,082). The GTX 770's competitive set is a full rung lower, which aligns with its lower percentile ranking.

The per-test splits reinforce the idea that these cards were tuned for different eras of API adoption. The GTX 690's OpenCL dominance suggests its dual-GPU array was well-served by compute abstraction layers that could distribute work across both GPUs. The GTX 770's Vulkan superiority indicates that its single, higher-clocked GK104 die responds better to modern low-level graphics APIs, which reward per-GPU efficiency over raw multi-GPU parallelism. In the database's recorded data, there is no test where the two cards land within single-digit percentage points of each other; every recorded result is a decisive win for one side.

The Verdict

The data points to a clear split based on workload and API preference. For users running OpenCL-centric tasks, the GTX 690 is the stronger choice. Its 10.8% lead in that test, combined with an average score that puts it in the 60th percentile overall, makes it the more capable compute card in legacy or OpenCL-heavy environments. It also holds its own against far newer hardware, sitting within 1% of the RTX 3070 in aggregate performance, which is a remarkable longevity signal for a 2012 flagship.

For Vulkan-based workloads, the GTX 770 is unequivocally the better card. Its 13% advantage in that test is the largest margin recorded between the two, and its Vulkan score of 19,166 is its single strongest result across any benchmark. The GTX 770 also carries a lower power envelope, with a 230 W TDP versus the GTX 690's 300 W, and requires only a 550 W suggested PSU compared to the GTX 690's 700 W. It is shorter as well, at 267 mm versus 279 mm, which matters for case compatibility.

The GTX 690's launch MSRP was 999 USD, while the GTX 770 launched at 399 USD. Both are end-of-life products today, but the database's measurements show that the GTX 690 still commands a higher performance tier in aggregate compute, while the GTX 770 offers the better modern-API experience and a more manageable power footprint. The choice depends entirely on which benchmark suite matters more to the user.

Architecture Differences

Both cards are built on the same GK104 chip, using NVIDIA's Kepler architecture, fabricated on TSMC's 28 nm process. Both have exactly 3,540 million transistors on a 294 mm² die, yielding a transistor density of 12.0 million per square millimeter. The shading unit count is identical at 1,536, as are the texture mapping units at 128 and the raster operation processors at 32. The memory configuration is also the same on paper: 2 GB of GDDR5 on a 256-bit bus. The differences emerge in clock speeds and memory timing.

The GTX 770 runs a base clock of 1046 MHz with a boost of 1085 MHz, while the GTX 690 operates at 915 MHz base and 1019 MHz boost. That gives the GTX 770 higher pixel rate (34.72 GPixel/s versus 32.61 GPixel/s) and texture rate (138.9 GTexel/s versus 130.4 GTexel/s), as well as higher FP32 compute at 3.333 TFLOPS versus 3.130 TFLOPS. The GTX 770 also clocks its memory at 1753 MHz (7 Gbps effective) for a bandwidth of 224.4 GB/s, while the GTX 690 runs at 1502 MHz (6 Gbps effective) for 192.3 GB/s. The GTX 690 makes up for these deficits with a second GK104 die, effectively doubling the shader, TMU, and ROP counts across the card, though the database records only the per-die specifications.

The GTX 690 is the older card, released in May 2012 as part of the GeForce 600 generation, while the GTX 770 arrived in May 2013 as a GeForce 700 product. Both support DirectX 12 (11_0), OpenGL 4.6, and Vulkan 1.2.175. Power delivery differs: the GTX 690 draws up to 300 W and requires two 8-pin connectors, while the GTX 770 draws 230 W with one 6-pin and one 8-pin connector. Display outputs also differ, with the GTX 690 offering three DVI ports and one mini-DisplayPort 1.2, while the GTX 770 offers two DVI ports, one HDMI 1.4a, and one DisplayPort 1.2. Both are dual-slot cards with identical height (111 mm) and width (38 mm), but the GTX 690 is longer at 279 mm versus the GTX 770's 267 mm.

The GTX 690's predecessor is the GeForce 500 series and its successor is the GeForce 700 series, while the GTX 770's predecessor is the GeForce 600 series and its successor is the GeForce 900 series. Neither card has ray tracing or tensor cores, as those features did not exist in the Kepler architecture.

FAQ

Q: Which card is faster in OpenCL workloads?

A: The GTX 690, with a Geekbench OpenCL score of 17,399 versus the GTX 770's 15,707, a 10.8% advantage.

Q: Which card wins in Vulkan benchmarks?

A: The GTX 770, scoring 19,166 against the GTX 690's 16,675, a 13% margin that is the largest recorded difference between the two.

Q: Do both cards use the same GPU chip?

A: Yes, both use the GK104 chip with 3,540 million transistors on a 294 mm² die manufactured on TSMC's 28 nm process, and both have 1,536 shading units, 128 TMUs, and 32 ROPs.

Q: How do the memory bandwidths compare?

A: The GTX 770 has higher bandwidth at 224.4 GB/s, driven by its 1753 MHz memory clock, while the GTX 690 offers 192.3 GB/s from a 1502 MHz memory clock. Both use 2 GB of GDDR5 on a 256-bit bus.

Q: What is the performance gap relative to modern GPUs?

A: The GTX 690's average score of 17,037 is within 1% of the RTX 3070's 17,208, while the GTX 770's average of 14,747 is 0.4% ahead of the RX 5500 XT's 14,692.

Q: Which card has lower power requirements?

A: The GTX 770, with a 230 W TDP and a suggested 550 W PSU, compared to the GTX 690's 300 W TDP and 700 W suggested PSU. The GTX 770 also uses one 6-pin and one 8-pin connector, while the GTX 690 requires two 8-pin connectors.

Where Each One Wins

The GTX 690 wins in aggregate compute performance. Its average benchmark score of 17,037 places it at the 60th percentile of all GPUs, and its OpenCL result of 17,399 is the standout figure. It is the card to pick for OpenCL-heavy applications, where its 10.8% lead over the GTX 770 is decisive. Its near-parity with the RTX 3070 in average score, within 1%, also makes it the stronger choice for users who want maximum raw compute from an older flagship. The trade-off is clear: a 300 W TDP, a 700 W PSU recommendation, a longer 279 mm PCB, and a dual-GPU design that does not scale well under Vulkan.

The GTX 770 wins in modern API performance. Its Vulkan score of 19,166 is its best result in any test and beats the GTX 690 by 13%. It also offers higher clock speeds across the board, with a base of 1046 MHz versus 915 MHz, and higher memory bandwidth at 224.4 GB/s versus 192.3 GB/s. Its pixel rate of 34.72 GPixel/s and texture rate of 138.9 GTexel/s outpace the GTX 690's 32.61 GPixel/s and 130.4 GTexel/s. For users prioritizing Vulkan-based games or applications, the GTX 770 is the better fit, and it does so with a lower 230 W TDP, a 550 W PSU suggestion, and a shorter 267 mm length. Its average score of 14,747 sits at the 57th percentile, slightly below the GTX 690, but its efficiency and modern API support make it the more balanced daily driver.

In short, the GTX 690 is the compute specialist with higher aggregate scores and a higher overall percentile, while the GTX 770 is the modern-API specialist with better clocks, bandwidth, and power efficiency. Neither card dominates the other across all recorded metrics; the database shows a clean one-to-one split that leaves the final decision to the user's workload priorities.