AMD Radeon R7 M360 vs NVIDIA GeForce GTX 765M Comparison

The NVIDIA GeForce GTX 765M and AMD Radeon R7 M360 are both end-of-life mobile GPUs from the 28 nm era, but the recorded data shows they were never really in the same weight class. The GTX 765M wins both head-to-head benchmarks, holds a higher percentile ranking against all GPUs in the database, and doubles the R7 M360 on raw compute and memory bandwidth. The R7 M360 counters with newer API feature levels and a much smaller die, which matters for the thin laptops it was designed for. What follows is a step-by-step walk through every number that separates them.

The Verdict

The database favors the GTX 765M in every performance-oriented comparison. It leads the Geekbench OpenCL test by 54.7% and the Geekbench Vulkan test by 28.5%, with an average benchmark score of 5501 against the R7 M360's 4931. Its percentile rank across all GPUs is 32 versus 29 for the AMD part.

Pick the GTX 765M if GPU throughput is the priority. It has 768 shading units against 384, 64 texture units against 24, 16 ROPs against 8, and 1,325.6 GFLOPS of FP32 compute against 864.0 GFLOPS. Memory bandwidth is 64.13 GB/s versus 14.40 GB/s, a 4.5x advantage rooted in GDDR5 on a 128-bit bus versus DDR3 on a 64-bit bus.

Pick the R7 M360 only if feature support or physical integration outweigh raw speed. It reports DirectX 12 with the 12_0 feature level versus 11_0 on the NVIDIA card, and it connects over a standard PCIe 3.0 x8 interface rather than an MXM module. Neither card has a listed launch MSRP, so the comparison here rests entirely on specifications and measured results.

FAQ

Q: Which GPU is faster overall?

A: The GTX 765M. It won both head-to-head tests in the database, taking Geekbench OpenCL 7176 to 4638 (54.7% faster) and Geekbench Vulkan 6714 to 5223 (28.5% faster).

Q: How do the two compare against the broader GPU landscape?

A: The GTX 765M sits in the 32nd percentile of all GPUs in the database, the R7 M360 in the 29th. Both are entry-level parts by modern standards.

Q: Which GPU has the stronger memory subsystem?

A: The GTX 765M by a wide margin. Its 2 GB of GDDR5 on a 128-bit bus delivers 64.13 GB/s, while the R7 M360's 2 GB of DDR3 on a 64-bit bus delivers 14.40 GB/s.

Q: Is either GPU still current?

A: No. Both are marked end-of-life in the database. The GTX 765M was released in May 2013, the R7 M360 in May 2015.

Q: Which rivals sit closest to each card?

A: The GTX 765M's nearest rivals are the GeForce MX130 (-0.1%), Radeon R7 M440 (0.3%), Quadro M4000 (0.6%), and FirePro M4000 (-0.7%). The R7 M360 clusters around the Radeon R7 M265 (0%), FirePro W5130M (0.6%), GeForce RTX 5060 Ti 8 GB (0.6%), and GeForce GTS 450 (0.8%).

Q: Which card has better DirectX feature support?

A: The R7 M360. It reports DirectX 12 at feature level 12_0, while the GTX 765M reports DirectX 12 at feature level 11_0. OpenGL is even at 4.6 on both.

Architecture Differences

Both GPUs were fabricated at TSMC on a 28 nm process, but that is where the architectural similarity ends. The GTX 765M uses NVIDIA's Kepler architecture on the GK106 chip, part of the GeForce 700M generation. The R7 M360 uses AMD's GCN 3.0 architecture on the "Meso" chip, part of the R7 M300 family.

The Kepler part is physically much larger: 2,540 million transistors on a 221 mm² die, giving a density of 11.5M transistors per mm². The Meso part packs 1,550 million transistors into 125 mm², a higher density of 12.4M per mm². The AMD design does more with less silicon, but the NVIDIA design simply has more hardware, and that shows up everywhere in the results.

The resource counts tell the story. The GTX 765M carries 768 shading units, 64 TMUs, and 16 ROPs. The R7 M360 carries 384 shading units, 24 TMUs, and 8 ROPs, exactly half or less of the NVIDIA allocation in every category. Pixel output is 13.81 GPixel/s versus 9.000 GPixel/s, and texture output is 55.23 GTexel/s versus 27.00 GTexel/s.

Clocking strategies differ. The GTX 765M runs a base clock of 797 MHz with a boost of 863 MHz. The R7 M360 clocks higher at 1100 MHz base and 1125 MHz boost, a compensating tactic that only partially offsets its smaller engine. Neither card has RT cores or tensor cores.

Specification Differences

• Release date: GTX 765M on May 29, 2013; R7 M360 on May 4, 2015.
• Memory type and bus: 2 GB GDDR5 on a 128-bit bus (64.13 GB/s) versus 2 GB DDR3 on a 64-bit bus (14.40 GB/s). Memory runs at 1002 MHz (4 Gbps effective) on the NVIDIA card, 900 MHz (1800 Mbps effective) on the AMD card.
• Shading units: 768 versus 384.
• Texture units: 64 versus 24.
• ROPs: 16 versus 8.
• FP32 compute: 1,325.6 GFLOPS versus 864.0 GFLOPS. The R7 M360 also lists FP16 at 864.0 GFLOPS (1:1 with FP32); no FP16 figure is recorded for the GTX 765M.
• Clocks: 797/863 MHz base/boost versus 1100/1125 MHz.
• Pixel rate: 13.81 GPixel/s versus 9.000 GPixel/s.
• Texture rate: 55.23 GTexel/s versus 27.00 GTexel/s.
• Interface: MXM-B (3.0) module versus PCIe 3.0 x8.
• Power: the GTX 765M lists a TDP of 75 W and no power connectors; no TDP is recorded for the R7 M360.
• APIs: DirectX 12 (11_0) and Vulkan 1.2.175 on the GTX 765M; DirectX 12 (12_0) and Vulkan 1.2.170 on the R7 M360. OpenGL is 4.6 on both.
• Lineage: the GTX 765M succeeds the GeForce 600M series and precedes the 800M series; the R7 M360 follows the "Solar System" generation and precedes Polaris Mobile.
• Transistors and die: 2,540 million on 221 mm² versus 1,550 million on 125 mm².

Head-to-Head Benchmarks

Only two tests overlap between the cards, and the GTX 765M won both, for a 2-0 record.

The largest gap is in Geekbench OpenCL, a general GPU compute test. The GTX 765M scored 7176 against 4638 for the R7 M360, a 54.7% lead. That margin tracks closely with the hardware gaps: double the shading units, double the ROPs, and roughly 4.5x the memory bandwidth. OpenCL workloads stress sustained throughput, and the Kepler card's wider engine plus GDDR5 subsystem simply moves far more data.

Geekbench Vulkan is closer. The GTX 765M scored 6714 against 5223, a 28.5% lead. The R7 M360's relatively stronger showing here is consistent with its GCN 3.0 architecture, which was designed with modern graphics APIs in mind, and its DirectX 12_0 feature level versus 11_0 on the NVIDIA card suggests a more complete modern API implementation even at lower throughput.

Context from each card's rival cluster reinforces the standings. The GTX 765M's average score of 5501 sits within a percentage point of the GeForce MX130, Radeon R7 M440, Quadro M4000, and FirePro M4000. The R7 M360's 4931 average puts it dead level with the Radeon R7 M265 and within 0.8% of the FirePro W5130M, GeForce RTX 5060 Ti 8 GB, and GTS 450 in the recorded data. Both cards occupy the same general performance neighborhood, but the NVIDIA part sits consistently above it.

Where Each One Wins

The GTX 765M wins on:

• Raw compute: 1,325.6 GFLOPS FP32 versus 864.0 GFLOPS, confirmed by a 54.7% OpenCL lead.
• Graphics throughput: higher pixel rate (13.81 versus 9.000 GPixel/s) and texture rate (55.23 versus 27.00 GTexel/s), plus double the shading units, TMUs, and ROPs.
• Memory-heavy workloads: 64.13 GB/s of GDDR5 bandwidth versus 14.40 GB/s of DDR3 on a narrow bus.
• Vulkan performance despite an older architecture, leading 6714 to 5223.

The R7 M360 wins on:

• Modern feature support: DirectX 12 at feature level 12_0 versus 11_0, with a comparable Vulkan version (1.2.170 versus 1.2.175).
• Die efficiency: a 125 mm² die versus 221 mm², with higher transistor density (12.4M versus 11.5M per mm²).
• FP16 capability: 864.0 GFLOPS at a 1:1 ratio with FP32, a feature not recorded on the Kepler card.
• Standard integration: a PCIe 3.0 x8 interface rather than a proprietary MXM module, and a listed TDP is absent, while the GTX 765M carries a 75 W rating.

In short, the data paints the GTX 765M as the performance pick of this pair, and the R7 M360 as the smaller, newer-feature card for scenarios where throughput is secondary.