AMD Radeon R7 350 vs NVIDIA GeForce GTX 750 Comparison

The AMD Radeon R7 350 and NVIDIA GeForce GTX 750 are both end-of-life, single-slot cards from the 28 nm era, but the benchmark data shows they are not equals. The GTX 750 wins both head-to-head tests decisively, and the specification sheets reveal why: NVIDIA’s Maxwell architecture simply extracts more performance from the same power envelope and memory configuration. The R7 350 does have a 2 GB frame buffer versus the GTX 750’s 1024 MB, but that advantage does not translate into a single benchmark victory.

Head-to-Head Benchmarks

The data is unambiguous: the NVIDIA GeForce GTX 750 outperforms the AMD Radeon R7 350 in every recorded test. In Geekbench OpenCL, the GTX 750 scores 9315 against the R7 350’s 7792, a delta of -16.3% for the AMD card. That is a substantial margin — roughly one-sixth less compute throughput in a workload that stresses raw shader and memory performance. The Vulkan test tells a similar story, with the GTX 750 posting 8078 versus 7057 for the R7 350, a -12.6% gap. While the Vulkan deficit is slightly smaller, it still represents a clear, consistent advantage for the NVIDIA card across both API environments.

Looking at the broader benchmark context, the R7 350’s average score of 7425 places it in the 40th percentile of all GPUs, and its nearest rivals include the Intel UHD Graphics 750 (7441, -0.2%) and AMD Radeon HD 8850M (7447, -0.3%). The GTX 750, by contrast, averages 7222, also in the 40th percentile, but its nearest rivals are the NVIDIA GeForce GTX 560 SE (7171, +0.7%) and Intel Iris Pro Graphics P580 (7170, +0.7%). Notably, the GTX 750’s average is dragged down by its Geekbench Metal score of 4274, a test the R7 350 did not take; in the two shared tests, the NVIDIA card leads by double digits.

The individual specification rates align with these results. The GTX 750’s pixel rate of 17.36 GPixel/s and texture rate of 34.72 GTexel/s exceed the R7 350’s 12.80 GPixel/s and 25.60 GTexel/s. Its FP32 throughput of 1,111.0 GFLOPS is roughly 36% higher than the R7 350’s 819.2 GFLOPS. Memory bandwidth also favors NVIDIA: 80.19 GB/s versus 72.00 GB/s, despite both using a 128-bit bus and GDDR5. The GTX 750’s effective memory speed is 5 Gbps, while the R7 350 operates at 4.5 Gbps, and its base clock of 1020 MHz with a boost of 1085 MHz gives it a frequency advantage that the R7 350’s datasheet does not even list.

FAQ

Q: Which card is faster in OpenCL workloads?

A: The NVIDIA GeForce GTX 750 scores 9315 in Geekbench OpenCL, which is 16.3% higher than the AMD Radeon R7 350’s 7792. This is the largest performance gap between the two cards in any test.

Q: Does the R7 350’s 2 GB memory give it any advantage?

A: The data does not show one. The R7 350 has 2 GB of GDDR5 versus the GTX 750’s 1024 MB, but the GTX 750 still wins both head-to-head benchmarks. The NVIDIA card also has higher memory bandwidth (80.19 GB/s vs 72.00 GB/s), so the larger frame buffer does not compensate for slower effective memory speed.

Q: How do these cards compare in Vulkan performance?

A: The GTX 750 scores 8078 in Geekbench Vulkan, which is 12.6% ahead of the R7 350’s 7057. The NVIDIA card also lists Vulkan 1.4 API support, while the R7 350 supports Vulkan 1.2.170.

Q: What is the average benchmark score for each card?

A: The R7 350 has an average benchmark score of 7425, while the GTX 750 averages 7222. However, the GTX 750’s average includes a lower Metal score of 4274, which the R7 350 does not have; in shared tests, the GTX 750 is consistently ahead.

Q: Are these cards comparable in power consumption?

A: Yes, both have a 55 W TDP and a suggested PSU rating of 250 W. Neither requires power connectors, and both are single-slot designs, making them drop-in replacements in similar low-power systems.

Q: Which card has better API support?

A: The GTX 750 supports Vulkan 1.4 and DirectX 12 (11_0), while the R7 350 supports Vulkan 1.2.170 and DirectX 12 (11_1). Both support OpenGL 4.6. The R7 350 has a slightly higher DirectX feature level, but the GTX 750 has a newer Vulkan version.

The Verdict

The choice is straightforward based on the data: the NVIDIA GeForce GTX 750 is the superior performer. It wins both head-to-head benchmarks with margins of 16.3% and 12.6%, and it does so with higher pixel, texture, and FP32 rates. The R7 350’s only hardware advantage is its 2 GB memory capacity, which does not produce a single win in any test. For a builder choosing between these two end-of-life cards, the GTX 750 delivers more compute performance per watt — both are 55 W — and its 145 mm length makes it easier to fit in compact cases than the R7 350’s 168 mm. The R7 350’s launch MSRP is not recorded, while the GTX 750 launched at 119 USD, but the performance gap alone justifies the NVIDIA pick.

There is one caveat: the R7 350 supports DirectX 12 (11_1) versus the GTX 750’s DirectX 12 (11_0), and it has a DisplayPort 1.2 output alongside HDMI 1.4a, whereas the GTX 750 offers two DVI outputs and a mini-HDMI 1.4a. If legacy display connectivity or a slightly higher DirectX feature level matters more than raw performance, the R7 350 has a niche appeal. Otherwise, the GTX 750 is the clear winner in every benchmark that matters.

Specification Differences

The two cards share several core specs: both use 28 nm TSMC fabrication, 512 shading units, 32 TMUs, 16 ROPs, a 128-bit memory bus, and GDDR5 memory. They also have identical 55 W TDPs, single-slot designs, no power connectors, a 250 W suggested PSU, and PCIe 3.0 x16 interfaces. The differences are where the performance gap originates.

The GTX 750 has explicit clock speeds: 1020 MHz base and 1085 MHz boost, while the R7 350 lists no base or boost clocks. Memory speed favors NVIDIA at 1253 MHz (5 Gbps effective) versus 1125 MHz (4.5 Gbps effective). Memory capacity differs: 1024 MB on the GTX 750 versus 2 GB on the R7 350. Bandwidth follows the clock speed, with the GTX 750 at 80.19 GB/s and the R7 350 at 72.00 GB/s. Pixel and texture rates are higher on the GTX 750 (17.36 GPixel/s and 34.72 GTexel/s) versus the R7 350 (12.80 GPixel/s and 25.60 GTexel/s). FP32 performance is 1,111.0 GFLOPS for NVIDIA versus 819.2 GFLOPS for AMD.

Physical dimensions differ: the GTX 750 is 145 mm (5.7 inches) long, while the R7 350 is 168 mm (6.6 inches). Display outputs are distinct — the R7 350 has 1x DVI, 1x HDMI 1.4a, and 1x DisplayPort 1.2; the GTX 750 has 2x DVI and 1x mini-HDMI 1.4a. API support also varies: the R7 350 offers DirectX 12 (11_1) and Vulkan 1.2.170, whereas the GTX 750 offers DirectX 12 (11_0) and Vulkan 1.4. Release dates are far apart: the R7 350 launched on 2016-07-05, while the GTX 750 arrived on 2014-02-17. The GTX 750’s launch MSRP was 119 USD; the R7 350 has no recorded launch MSRP.

Architecture Differences

The architectural split is generational. The R7 350 uses GCN 1.0 (chip codename Cape Verde) from the Pirate Islands (R7 300) generation, while the GTX 750 uses Maxwell (chip codename GM107) from the GeForce 700 generation. Both are built on 28 nm at TSMC, but the transistor counts differ: NVIDIA packs 1,870 million transistors into a 148 mm² die (12.6M / mm²), while AMD uses 1,500 million on a 123 mm² die (12.2M / mm²). The GTX 750 has more transistors and a larger die, which correlates with its higher throughput rates.

The R7 350’s predecessor is Volcanic Islands, and its successor is Arctic Islands; the GTX 750’s predecessor is GeForce 600, and its successor is GeForce 900. This places the two cards in different market eras — the GTX 750 is a first-gen Maxwell part, while the R7 350 is a later GCN refresh. The API support reflects this: Maxwell’s Vulkan 1.4 support is more current than GCN 1.0’s Vulkan 1.2.170, though AMD’s DirectX 12 (11_1) feature level is one notch higher than NVIDIA’s (11_0). Neither card has RT cores or tensor cores, and neither lists FP16 support, so both are purely rasterization-focused designs.

Where Each One Wins

The NVIDIA GeForce GTX 750 wins every benchmark category where both cards were tested: OpenCL (9315 vs 7792) and Vulkan (8078 vs 7057). It also wins on every raw throughput metric — pixel rate, texture rate, FP32, and memory bandwidth. The GTX 750 is the card for anyone prioritizing compute performance in OpenCL or Vulkan applications, and its shorter 145 mm length makes it the better fit for small form factor builds.

The AMD Radeon R7 350 wins in exactly one meaningful area: memory capacity, with 2 GB versus 1024 MB. This could matter for workloads that exceed 1 GB of frame buffer, though the benchmark data does not show any test where this translates into a victory. The R7 350 also has a DisplayPort 1.2 output, which the GTX 750 lacks, and its DirectX 12 (11_1) feature level is slightly higher. For a builder with a display that requires DisplayPort or an application that specifically needs DirectX 11_1 features, the R7 350 has a narrow use case. Otherwise, the data is one-sided: the GTX 750 is the faster, more capable card in every measured scenario.