NVIDIA GeForce GTX 1650 SUPER vs NVIDIA GeForce GTX 950 Comparison

Head-to-Head Benchmarks

The recorded data shows two shared benchmark tests between the NVIDIA GeForce GTX 950 and the NVIDIA GeForce GTX 1650 SUPER: Geekbench OpenCL and Geekbench Vulkan. In both, the GTX 1650 SUPER is the clear winner, but the magnitude of the margin tells the real story.

In Geekbench OpenCL, the GTX 950 scores 15,662 points. The GTX 1650 SUPER scores 43,875 points, which is a 64.3% advantage. That is not a marginal improvement; it is a generational leap in raw compute throughput. The GTX 1650 SUPER delivers nearly three times the OpenCL score of its older counterpart.

The Vulkan test shows an even wider gap. The GTX 950 records 16,734 points, while the GTX 1650 SUPER reaches 50,519 points. The delta here is 66.9% in favor of the newer card. This suggests that the GTX 1650 SUPER is particularly well-suited to modern graphics APIs, whereas the GTX 950, despite its Maxwell architecture supporting Vulkan 1.4, falls far behind in practical API-level performance.

Neither card wins a benchmark outright in the GTX 950's favor. The head-to-head record is 0 wins for the GTX 950 and 2 wins for the GTX 1650 SUPER. However, context from the database's overall scores moderates this picture. The GTX 950's average benchmark score across all recorded tests is 13,189, while the GTX 1650 SUPER's average is 11,047. This is an unusual inversion: the older card has a higher average score despite losing both shared tests. The explanation lies in the different benchmark suites each card was tested with. The GTX 1650 SUPER includes a wide range of PassMark tests, including DirectX 9, 10, 11, and 12 subtests, plus G2D and GPU compute, which pull its average down. The GTX 950's average is based on a smaller set of Geekbench tests, all of which are compute-oriented.

Looking at the nearest rivals in the database, the GTX 950 sits at the 53rd percentile of all GPUs, while the GTX 1650 SUPER sits at the 50th. The GTX 950's closest competitor is the NVIDIA GeForce GTX 480, which averages 13,300 points, a 0.8% difference. The AMD Radeon Pro 555X is 1% ahead, and the AMD FirePro M6100 is 1.2% ahead. The GTX 1650 SUPER's nearest rival is the AMD Radeon RX 550 at 11,075 points, which is 0.2% ahead. Interestingly, the GTX 1650 SUPER also sits near the NVIDIA RTX PRO 6000D Blackwell Max-Q and the RTX PRO 6000 Blackwell Max-Q, both at 11,088 points, 0.4% ahead, and the NVIDIA GeForce MX350 at 10,883 points, 1.5% behind. This places the GTX 1650 SUPER in a different performance tier entirely, despite the lower average score.

Where Each One Wins

The GTX 1650 SUPER wins decisively in compute-heavy workloads that leverage OpenCL and Vulkan. The Geekbench OpenCL score of 43,875 versus 15,662 demonstrates a 64.3% lead, which suggests strong performance in general-purpose GPU compute tasks such as video encoding, physics simulations, and data-parallel workloads. The Vulkan score of 50,519 versus 16,734, a 66.9% lead, indicates that the GTX 1650 SUPER is far better equipped for modern games and applications that use Vulkan as their rendering API. This is the card to choose for anyone running contemporary titles or compute applications that take advantage of these interfaces.

The GTX 950, by contrast, has no recorded wins in the shared benchmarks. However, its higher average benchmark score of 13,189 versus 11,047 suggests that in the specific tests it was subjected to, it holds up well against its own peer group. The GTX 950's nearest rivals all score within 1.2% of its average, meaning it is competitive with other mid-range cards from its era. The GTX 1650 SUPER's average score of 11,047 places it just 0.2% behind the AMD Radeon RX 550, so within its own cohort, it is also competitive, though not dominant.

The use-case split is therefore clear: the GTX 1650 SUPER is the better choice for modern API-based workloads and compute tasks. The GTX 950, while older, still holds its own in the legacy benchmark environment where it was measured, but it cannot match the newer card in the tests where both were directly compared.

Architecture Differences

The two cards come from different architectural generations. The GTX 950 is built on the GM206 chip using Maxwell 2.0 architecture, part of the GeForce 900 generation. The GTX 1650 SUPER uses the TU116 chip with Turing architecture, part of the GeForce 16 generation. This is a fundamental shift in GPU design.

The manufacturing process also differs. The GTX 950 uses a 28 nm process at TSMC, while the GTX 1650 SUPER uses a 12 nm process, also at TSMC. This process shrink allows for a significantly higher transistor count. The GTX 950 has 2,940 million transistors on a 228 mm² die, giving a transistor density of 12.9 million per square millimeter. The GTX 1650 SUPER packs 6,600 million transistors onto a 284 mm² die, achieving a density of 23.2 million per square millimeter. The newer card has more than double the transistor count and nearly double the density, which explains its substantial performance advantage in the benchmarks.

Shader resources also scale up. The GTX 950 has 768 shading units, 48 texture mapping units, and 32 raster output units. The GTX 1650 SUPER has 1,280 shading units, 80 texture mapping units, and still 32 raster output units. The increase in shading units and TMUs directly contributes to the higher texture and pixel rates. The GTX 950's texture rate is 57.02 GTexel/s, while the GTX 1650 SUPER achieves 138.0 GTexel/s, more than double. The pixel rate rises from 38.02 GPixel/s to 55.20 GPixel/s.

Neither card includes ray tracing cores or tensor cores, so both are limited to traditional rasterization and compute workloads. Both support DirectX 12 (12_1), OpenGL 4.6, and Vulkan 1.4, so API compatibility is not a differentiator. The GTX 1650 SUPER does add FP16 support with a 2:1 ratio, offering 8.832 TFLOPS of FP16 compute, while the GTX 950 has no recorded FP16 capability.

Specification Differences

The most obvious difference is memory. The GTX 950 has 2 GB of GDDR5 memory on a 128-bit bus, with a bandwidth of 105.8 GB/s. The GTX 1650 SUPER has 4 GB of GDDR6 memory on the same 128-bit bus, but bandwidth jumps to 192.0 GB/s. This doubling of memory capacity and near-doubling of bandwidth is critical for modern games and larger textures.

Clock speeds differ substantially. The GTX 950 runs at a base clock of 1024 MHz and a boost clock of 1188 MHz. The GTX 1650 SUPER runs at 1530 MHz base and 1725 MHz boost, a significant increase. Memory clocks also differ: the GTX 950's memory runs at 1653 MHz, which translates to 6.6 Gbps effective, while the GTX 1650 SUPER's memory runs at 1500 MHz, but with a 12 Gbps effective rate due to GDDR6's architecture.

Compute throughput is vastly different. The GTX 950 delivers 1.825 TFLOPS of FP32 performance, while the GTX 1650 SUPER delivers 4.416 TFLOPS, a 142% increase. The GTX 1650 SUPER also has FP16 performance of 8.832 TFLOPS, which the GTX 950 does not record.

Power requirements are close but not identical. The GTX 950 has a TDP of 90 W and a suggested PSU of 250 W. The GTX 1650 SUPER has a TDP of 100 W and a suggested PSU of 300 W. Both use a single 6-pin power connector and are dual-slot cards. Physical dimensions differ slightly: the GTX 950 is 202 mm (8 inches) long, while the GTX 1650 SUPER is 229 mm (9 inches) long, with a height of 111 mm (4.4 inches) and a width of 35 mm (1.4 inches). The GTX 950 has no recorded height or width.

Display outputs differ. The GTX 950 offers 1x DVI, 1x HDMI 2.0, and 3x DisplayPort 1.2. The GTX 1650 SUPER offers 1x DVI, 1x HDMI 2.0, and 1x DisplayPort 1.4a. The newer card supports a newer DisplayPort standard but has fewer DisplayPort outputs.

Both cards have a launch MSRP of 159 USD. The GTX 950 was released on 2015-08-19, and the GTX 1650 SUPER on 2019-11-21. Both are end-of-life products. The GTX 950's predecessor is the GeForce 700 series and its successor is the GeForce 10 series. The GTX 1650 SUPER's predecessor is the GeForce 10 series and its successor is the GeForce 20 series.

FAQ

Q: Which card performs better in Geekbench OpenCL?

A: The GTX 1650 SUPER scores 43,875 versus the GTX 950's 15,662, a 64.3% advantage.

Q: Is the GTX 950 better in any shared benchmark?

A: No. The GTX 1650 SUPER wins both shared tests, Geekbench OpenCL and Geekbench Vulkan, with the GTX 950 recording 0 wins.

Q: How does the GTX 1650 SUPER's Vulkan performance compare?

A: It scores 50,519 in Geekbench Vulkan, which is 66.9% higher than the GTX 950's 16,734.

Q: What is the memory configuration difference?

A: The GTX 950 has 2 GB of GDDR5 with 105.8 GB/s bandwidth, while the GTX 1650 SUPER has 4 GB of GDDR6 with 192.0 GB/s bandwidth. Both use a 128-bit bus.

Q: Are both cards still in production?

A: No. Both are end-of-life products. The GTX 950 was released on 2015-08-19, and the GTX 1650 SUPER on 2019-11-21.

Q: Do either cards support ray tracing?

A: No. Neither the GTX 950 nor the GTX 1650 SUPER includes ray tracing cores or tensor cores.

The Verdict

The data points to one clear conclusion: the GTX 1650 SUPER is the superior card for anyone running modern workloads. In the two shared benchmarks, it beats the GTX 950 by 64.3% in OpenCL and 66.9% in Vulkan. Its 4 GB of GDDR6 memory, 192.0 GB/s bandwidth, and 4.416 TFLOPS of FP32 performance give it a commanding lead in both compute and rendering tasks.

The GTX 950, despite its higher average benchmark score of 13,189, cannot compete in the tests where both cards were measured. That higher average reflects a different benchmark suite, not superior performance. Its 2 GB of GDDR5 memory and 1.825 TFLOPS of FP32 throughput place it firmly in an older performance class.

For users with legacy applications that do not leverage Vulkan or OpenCL extensively, the GTX 950 remains functional, and its 53rd percentile ranking shows it is competitive with its own era's mid-range cards. But for anyone choosing between these two today, the GTX 1650 SUPER's 50th percentile ranking, despite a lower average score, is misleading; the head-to-head results are unambiguous. The GTX 1650 SUPER is the only rational pick for modern gaming and compute.