NVIDIA GeForce GTX 690 vs NVIDIA GeForce RTX 3050 OEM Comparison

FAQ

Q: How does the GeForce GTX 690 compare to the RTX 3050 OEM in overall average benchmark scores?

A: The GTX 690 has an average benchmark score of 17,037, while the RTX 3050 OEM records 15,199. Despite the lower average, the RTX 3050 OEM wins both head-to-head tests, indicating the average is skewed by the larger test set available for the newer card.

Q: Which card has the higher percentile ranking among all GPUs?

A: The GTX 690 sits at the 60th percentile, while the RTX 3050 OEM ranks at the 57th percentile. The GTX 690 places slightly higher in the overall distribution despite its older architecture.

Q: What are the closest rivals to each card according to the database?

A: The GTX 690's nearest rivals include the AMD Radeon HD 7970M (0.1% faster), AMD Radeon RX 7600 XT (0.3% faster), NVIDIA Tesla M4 (0.6% slower), and NVIDIA GeForce RTX 3070 (1% faster). The RTX 3050 OEM's nearest rivals include the AMD Radeon RX 7600 (0.2% faster), AMD Radeon 680M (0.5% faster), NVIDIA GeForce GTX 580 (0.5% faster), and NVIDIA GeForce RTX 2060 (0.6% faster).

Q: Which card shows the larger lead in the Geekbench OpenCL test?

A: The RTX 3050 OEM scores 60,740 versus 17,399 for the GTX 690, a delta of -71.4% from the perspective of the older card. This represents a roughly 3.5x advantage for the RTX 3050 OEM.

Q: What is the difference in memory capacity and type between the two cards?

A: The GTX 690 features 2 GB of GDDR5 memory on a 256-bit bus, while the RTX 3050 OEM offers 8 GB of GDDR6 memory on a 128-bit bus. The newer card has four times the capacity but half the bus width.

Q: Do both cards support the same DirectX version?

A: No. The GTX 690 supports DirectX 12 (11_0), while the RTX 3050 OEM supports DirectX 12 Ultimate (12_2). The newer card also lists Vulkan 1.4 support versus Vulkan 1.2.175 for the GTX 690.

Where Each One Wins

The RTX 3050 OEM wins decisively in both recorded head-to-head benchmark tests. In Geekbench OpenCL, it scores 60,740 versus 17,399 for the GTX 690, a 71.4% gap. In Geekbench Vulkan, the margin is similar: 57,103 against 16,675, a 70.8% difference. These are not narrow victories; the newer card delivers roughly three and a half times the compute throughput in both APIs.

The GTX 690, however, retains a statistical edge in percentile placement. Its 60th percentile versus the RTX 3050 OEM's 57th percentile indicates that, across the entire database of GPUs, the older dual-GPU card sits in a slightly higher position. This is notable given that the GTX 690's average score of 17,037 exceeds the RTX 3050 OEM's 15,199. The database's larger test set for the RTX 3050 OEM, which includes Passmark DirectX 9 through 12, G2D, G3D, and compute tests, drags its average down relative to its strong Geekbench results.

For legacy DirectX workloads, the GTX 690's Kepler architecture may hold an advantage in specific scenarios, but the recorded data does not include such tests for that card. The RTX 3050 OEM's Passmark scores are available: DirectX 9 at 137, DirectX 10 at 61, DirectX 11 at 86, DirectX 12 at 58, and G3D at 11,857. These figures suggest the newer card handles older APIs adequately but shines in modern compute-oriented tasks where its 8.087 TFLOPS FP32 throughput becomes relevant.

The GTX 690 wins on raw texture fill rate with 130.4 GTexel/s versus 126.4 GTexel/s for the RTX 3050 OEM. This is a narrow margin, under 3%, and reflects the GTX 690's 128 TMUs versus 72 on the newer card. In pixel throughput, the RTX 3050 OEM reverses the trend with 56.16 GPixel/s versus 32.61 GPixel/s, a 72% advantage.

Architecture Differences

The GTX 690 is built on the Kepler architecture using the GK104 chip, manufactured on TSMC's 28 nm process. It packs 3,540 million transistors into a 294 mm² die, yielding a transistor density of 12.0M per mm². The RTX 3050 OEM uses the Ampere architecture with the GA106 chip, fabricated by Samsung on an 8 nm process. This newer design crams 12,000 million transistors into a 276 mm² die, achieving a density of 43.5M per mm², which is over 3.6 times denser than the Kepler part.

The GTX 690 is a dual-GPU card, which explains its 300 W TDP and dual 8-pin power connectors. The RTX 3050 OEM is a single-chip solution with a 130 W TDP and a single 8-pin connector, less than half the power draw. The newer card also brings hardware ray tracing cores (18) and tensor cores (72), features completely absent from the Kepler architecture.

Memory architecture differs fundamentally. The GTX 690 uses 2 GB of GDDR5 on a 256-bit bus, delivering 192.3 GB/s bandwidth. The RTX 3050 OEM offers 8 GB of GDDR6 on a 128-bit bus, with 224.0 GB/s bandwidth. Despite the narrower bus, the newer card achieves 16% more bandwidth thanks to faster memory running at 14 Gbps effective versus 6 Gbps effective.

The RTX 3050 OEM supports PCIe 4.0 x8, while the GTX 690 runs on PCIe 3.0 x16. The newer card also supports DirectX 12 Ultimate (12_2) and Vulkan 1.4, whereas the GTX 690 lists DirectX 12 (11_0) and Vulkan 1.2.175. Both support OpenGL 4.6.

Specification Differences

The two cards differ across nearly every specification field. Process node: 28 nm versus 8 nm. Transistor count: 3,540 million versus 12,000 million. Die size: 294 mm² versus 276 mm². Transistor density: 12.0M versus 43.5M per mm².

Clock speeds: the GTX 690 runs at a base of 915 MHz with a boost of 1019 MHz, while the RTX 3050 OEM operates at 1515 MHz base and 1755 MHz boost. Memory clocks: 1502 MHz (6 Gbps effective) versus 1750 MHz (14 Gbps effective).

Memory specs: 2 GB GDDR5 on 256-bit versus 8 GB GDDR6 on 128-bit. Bandwidth: 192.3 GB/s versus 224.0 GB/s.

Compute units: the GTX 690 has 1536 shading units, 128 TMUs, and 32 ROPs. The RTX 3050 OEM has 2304 shading units, 72 TMUs, and 32 ROPs. The newer card adds 18 RT cores and 72 tensor cores, which the GTX 690 lacks.

Pixel rate: 32.61 GPixel/s versus 56.16 GPixel/s. Texture rate: 130.4 GTexel/s versus 126.4 GTexel/s. FP32: 3.130 TFLOPS versus 8.087 TFLOPS. The RTX 3050 OEM also lists FP16 at 8.087 TFLOPS (1:1), while the GTX 690 has no FP16 figure.

Power: 300 W versus 130 W TDP. Power connectors: 2x 8-pin versus 1x 8-pin. Suggested PSU: 700 W versus 300 W.

Physical dimensions: the GTX 690 is 279 mm long, 111 mm high, and 38 mm wide. The RTX 3050 OEM is 242 mm long and 112 mm high, with no width listed. Both are dual-slot cards.

Display outputs: the GTX 690 offers 3x DVI and 1x mini-DisplayPort 1.2. The RTX 3050 OEM provides 1x HDMI 2.1 and 3x DisplayPort 1.4a.

Release dates: May 2, 2012 for the GTX 690 versus January 3, 2022 for the RTX 3050 OEM. Both are end-of-life products.

Head-to-Head Benchmarks

The database records two head-to-head comparisons, both favoring the RTX 3050 OEM by wide margins.

In Geekbench OpenCL, the RTX 3050 OEM scores 60,740 against the GTX 690's 17,399. The delta is -71.4%, meaning the GTX 690 delivers only 28.6% of the newer card's performance. This gap reflects the fundamental architectural leap: the RTX 3050 OEM's 8.087 TFLOPS FP32 throughput is 2.6 times the GTX 690's 3.130 TFLOPS. The newer card also benefits from higher memory bandwidth (224.0 GB/s versus 192.3 GB/s) and faster boost clocks (1755 MHz versus 1019 MHz).

In Geekbench Vulkan, the results are similar. The RTX 3050 OEM scores 57,103, while the GTX 690 manages 16,675, a delta of -70.8%. The Vulkan gap is slightly narrower than OpenCL, but the pattern holds. The RTX 3050 OEM's Vulkan 1.4 support versus the GTX 690's Vulkan 1.2.175 likely contributes to its stronger showing in this API.

Looking at the GTX 690's nearest rivals provides context. It sits within 1% of the AMD Radeon HD 7970M (17,019), AMD Radeon RX 7600 XT (17,083), NVIDIA Tesla M4 (16,932), and NVIDIA GeForce RTX 3070 (17,208). These scores cluster tightly around 17,000, suggesting the GTX 690's dual-GPU design delivers consistent compute performance across a narrow band.

The RTX 3050 OEM's nearest rivals tell a different story. Its average score of 15,199 places it within 0.6% of the AMD Radeon RX 7600 (15,171), AMD Radeon 680M (15,270), NVIDIA GeForce GTX 580 (15,283), and NVIDIA GeForce RTX 2060 (15,290). The cluster around 15,200 is notably lower than the GTX 690's cluster around 17,000, yet the RTX 3050 OEM crushes the GTX 690 in the two Geekbench tests. This discrepancy arises because the RTX 3050 OEM's average includes nine benchmark entries, many of which (Passmark DirectX 9 through 12) score in the dozens or hundreds, pulling the average down.

The GTX 690's average of 17,037 comes from only two Geekbench tests, both in the 16,000 to 17,000 range. If the GTX 690 had Passmark results, its average would likely differ. The recorded data shows the RTX 3050 OEM's Passmark G3D score of 11,857, which is substantial but lower than its Geekbench scores.

The wins tally is 0 for the GTX 690 and 2 for the RTX 3050 OEM. No benchmark in the database favors the older card. The closest the GTX 690 comes to a win is in texture rate, where its 130.4 GTexel/s narrowly exceeds the RTX 3050 OEM's 126.4 GTexel/s, a difference of 3.2%. This advantage stems from the GTX 690's 128 TMUs versus 72, though the newer card's higher clocks partially compensate.

For pixel throughput, the RTX 3050 OEM leads by 72% (56.16 GPixel/s versus 32.61 GPixel/s), driven by its higher ROP efficiency and boost clock. The GTX 690's dual-GPU configuration does not double its pixel rate because each GPU has 32 ROPs, and the aggregate is 32.61 GPixel/s, matching a single Kepler GPU.

The FP32 gap is the most telling specification difference: 8.087 TFLOPS versus 3.130 TFLOPS. This 2.6x advantage in raw compute explains the Geekbench results, which heavily weight arithmetic throughput. The RTX 3050 OEM's FP16 capability at 8.087 TFLOPS (1:1) further extends its compute lead for mixed-precision workloads, a feature the GTX 690 lacks entirely.

In summary, the RTX 3050 OEM wins every recorded benchmark by a substantial margin, while the GTX 690 retains a slight edge in texture fill rate and a higher percentile ranking. The database's limited test set for the GTX 690 makes direct comparisons challenging, but the available evidence points decisively to the newer Ampere card as the superior compute platform.