AMD Radeon R9 M375X vs NVIDIA GeForce GTX 1650 Comparison

The AMD Radeon R9 M375X and NVIDIA GeForce GTX 1650 sit in different eras of GPU design, and the recorded data makes the gap between them plain. Both are end-of-life parts now, but one of them dominates every head-to-head measurement in the database. The R9 M375X is a 2015-era mobile Radeon built on GCN 1.0, while the GTX 1650 is a Turing-based desktop card from 2019, and the benchmark results reflect that generational divide in stark terms.

Head-to-Head Benchmarks

Only two tests exist where both GPUs have recorded results, and the GTX 1650 sweeps them both.

In Geekbench OpenCL, the GTX 1650 scores 29,629 against 8,273 for the R9 M375X. That is a 72.1 percent margin for NVIDIA. The Vulkan test widens the gap further: 33,042 versus 8,377, a 74.6 percent win for the GTX 1650. The R9 M375X takes zero head-to-head victories; the GTX 1650 takes both.

Those are not narrow margins. A delta of roughly three-quarters in both compute APIs means the GTX 1650 delivers close to four times the throughput of the R9 M375X in the measured workloads. The raw theoretical figures back this up: the GTX 1650 posts 2.984 TFLOPS of FP32 compute against 1,299.2 GFLOPS for the R9 M375X, and it also supports FP16 at 5.967 TFLOPS in a 2:1 ratio, a capability the AMD part simply lacks a recorded figure for.

Context from each card's rival set sharpens the picture. The R9 M375X's average benchmark score of 8,325 puts it alongside the NVIDIA Quadro K1200 (8,265, within 0.7 percent), the GeForce GTX 675MX (8,427, 1.2 percent ahead), the AMD Radeon 880M (8,436, 1.3 percent ahead), and the GeForce MX330 (8,458, 1.6 percent ahead). In other words, it trades blows with entry-level workstation cards and older mobile parts. The GTX 1650's average of 7,472 places it near the Radeon HD 8850M, Intel UHD Graphics 750, the Radeon R7 350, and the Intel Arc A310 (within 1 percent in every case). Note that the two averages come from different test mixes, so the head-to-head scores are the ones that matter, and they are unanimous.

Where Each One Wins

The GTX 1650 wins everywhere the database has direct measurements. OpenCL compute, Vulkan compute, pixel fill rate, texture throughput, memory bandwidth: every comparable recorded figure favors NVIDIA.

The scale of those hardware advantages is consistent. Pixel rate is 53.28 GPixel/s versus 16.24 GPixel/s, more than three times higher. Texture rate is 93.24 GTexel/s versus 40.60 GTexel/s. Memory bandwidth stands at 128.1 GB/s against 72.00 GB/s, roughly 78 percent more, despite both cards using a 128-bit GDDR5 bus, because the GTX 1650's memory runs at 8 Gbps effective compared to 4.5 Gbps effective on the R9 M375X. The GTX 1650 also doubles the frame buffer, 4 GB versus 2 GB, which matters for modern game asset sizes and higher texture settings.

For the R9 M375X, the honest summary is that the data shows no benchmark category where it wins. Its percentile placement against all GPUs in the database, 43 versus 40 for the GTX 1650, differs slightly, but those percentiles reflect different test distributions, not head-to-head performance. On the two tests both cards actually share, the AMD part is not competitive.

Architecture Differences

The two cards come from opposite ends of a technology transition. The R9 M375X uses the Tropo chip on GCN 1.0 architecture, fabricated at TSMC on a 28 nm process, with 1,500 million transistors packed into a 123 mm² die for a density of 12.2M transistors per mm². The GTX 1650 uses the TU117 chip on Turing architecture, built at TSMC on a 12 nm process, with 4,700 million transistors across a 200 mm² die at 23.5M per mm², nearly double the density.

The GTX 1650 belongs to the GeForce 16 generation and succeeds the GeForce 10 line; the R9 M375X belongs to AMD's R9 M300 generation, succeeded by Polaris Mobile. Neither card has RT cores or tensor cores listed in the database, so ray tracing and AI acceleration are absent from both spec sheets.

Shader resources also diverge sharply. The GTX 1650 carries 896 shading units, 56 TMUs, and 32 ROPs. The R9 M375X has 640 shading units, 40 TMUs, and 16 ROPs, half the raster output throughput. Clocks compound the difference: 1,485 MHz base and 1,665 MHz boost on the GTX 1650 versus 925 MHz base and 1,015 MHz boost on the AMD card.

Software support favors NVIDIA as well. The GTX 1650 supports DirectX 12 at feature level 12_1 and Vulkan 1.4, while the R9 M375X tops out at DirectX 12 (11_1) and Vulkan 1.2.170. Both support OpenGL 4.6. The GTX 1650 was released in April 2019; the R9 M375X in May 2015.

Form factor and power data reflect their intended homes. The GTX 1650 is a dual-slot desktop card measuring 229 mm long, 111 mm high, and 35 mm wide, with a 75 W TDP, no power connectors, and a suggested 250 W PSU. It offers one DVI, one HDMI 2.0, and one DisplayPort 1.4a output. The R9 M375X is a mobile part with no recorded TDP, slot width, dimensions, or display outputs in the database. Both connect over PCIe 3.0 x16.

FAQ

Q: Is the GTX 1650 faster than the R9 M375X in compute benchmarks?

A: Yes, decisively. The GTX 1650 scores 29,629 in Geekbench OpenCL and 33,042 in Geekbench Vulkan, versus 8,273 and 8,377 for the R9 M375X, wins of 72.1 and 74.6 percent respectively.

Q: Which card has more video memory?

A: The GTX 1650, with 4 GB of GDDR5 compared to 2 GB on the R9 M375X. Both use a 128-bit bus, but the GTX 1650's 128.1 GB/s bandwidth exceeds the R9 M375X's 72.00 GB/s.

Q: Do either of these cards support ray tracing hardware?

A: No. Neither GPU has RT cores or tensor cores listed in the database, so hardware ray tracing and dedicated AI acceleration are absent from both.

Q: What DirectX feature levels do they support?

A: The GTX 1650 supports DirectX 12 at feature level 12_1; the R9 M375X supports DirectX 12 at feature level 11_1. The GTX 1650 also supports the newer Vulkan 1.4 versus 1.2.170 on the AMD card.

Q: What was the GTX 1650's launch MSRP?

A: 149 USD at launch.

Q: Are both cards still in production?

A: No, both are marked end-of-life in the database. The R9 M375X launched in May 2015 and the GTX 1650 in April 2019.

The Verdict

Based strictly on the recorded data, there is no scenario in these results where the R9 M375X is the stronger pick. It loses both shared benchmarks by more than 70 percent, has half the memory, roughly half the shading hardware by every relevant metric, older API support, and a process node two full steps behind. Its rival cohort, the Quadro K1200, GTX 675MX, Radeon 880M, and MX330, confirms it as an entry-level part of its era.

The GTX 1650 is the clear choice for anything compute- or graphics-related measured here: nearly four times the OpenCL and Vulkan throughput, 53.28 versus 16.24 GPixel/s of pixel fill, 93.24 versus 40.60 GTexel/s of texture fill, and a 4 GB buffer for modern workloads. Its 75 W TDP, lack of power connectors, and 250 W suggested PSU also make it an easy drop-in for compact desktop builds. The R9 M375X's relevance is historical, a mobile GCN part from 2015 whose data is useful mainly for tracking generational progress.

Specification Differences

• Architecture: GCN 1.0 (R9 M375X) vs Turing (GTX 1650)
• Chip: Tropo vs TU117
• Generation: R9 M300 (Gem System) vs GeForce 16
• Process node: 28 nm vs 12 nm, both at TSMC
• Transistors: 1,500 million vs 4,700 million
• Die size: 123 mm² vs 200 mm²
• Transistor density: 12.2M/mm² vs 23.5M/mm²
• Base / boost clock: 925 / 1015 MHz vs 1485 / 1665 MHz
• Memory size: 2 GB vs 4 GB (both GDDR5, 128-bit)
• Memory speed: 4.5 Gbps effective vs 8 Gbps effective
• Memory bandwidth: 72.00 GB/s vs 128.1 GB/s
• Shading units: 640 vs 896
• TMUs: 40 vs 56
• ROPs: 16 vs 32
• Pixel rate: 16.24 vs 53.28 GPixel/s
• Texture rate: 40.60 vs 93.24 GTexel/s
• FP32: 1,299.2 GFLOPS vs 2.984 TFLOPS
• FP16: not recorded vs 5.967 TFLOPS (2:1)
• DirectX: 12 (11_1) vs 12 (12_1)
• Vulkan: 1.2.170 vs 1.4
• TDP: not recorded vs 75 W
• Slot width: not recorded vs dual-slot
• Power connectors: not recorded vs none
• Suggested PSU: not recorded vs 250 W
• Display outputs: not recorded vs 1x DVI, 1x HDMI 2.0, 1x DisplayPort 1.4a
• Dimensions: not recorded vs 229 x 111 x 35 mm
• Release date: May 2015 vs April 2019
• Predecessor / successor: Solar System / Polaris Mobile vs GeForce 10 / GeForce 20
• Launch MSRP: not recorded vs 149 USD
• Average benchmark score: 8,325 vs 7,472 (different test mixes; head-to-head results favor the GTX 1650 in both shared tests)