AMD Radeon R5 M320 vs NVIDIA GeForce GTX 760M Comparison

Head-to-Head Benchmarks

The benchmark data shows a clear overall winner in this pairing, with the NVIDIA GeForce GTX 760M taking both recorded tests. In Geekbench OpenCL, the GTX 760M scores 5604 against the AMD Radeon R5 M320's 5051, a 10.9% lead. The gap widens in Geekbench Vulkan, where the GTX 760M records 4868 versus 4262, a 14.2% advantage. Across both tests, the GTX 760M wins 2 tests out of 2; the R5 M320 does not win either.

The OpenCL result of 5604 places the GTX 760M in a competitive position relative to its closest rivals. It sits 0.5% ahead of the NVIDIA Quadro 4000M, whose average score is 5211, and 1.5% ahead of the AMD Radeon R7 M320's nearest rival, the AMD Radeon RX 9060 XT 16 GB, which averages 4657. The R5 M320's OpenCL score of 5051 places it 0.5% behind the NVIDIA Quadro 4000M and 1.5% ahead of the AMD Radeon R7 M260X. The Vulkan result for the R5 M320, 4262, is 0.8% ahead of the NVIDIA Quadro M3000M, which averages 4621. For reference, the GTX 760M's Vulkan score of 4868 is 0.6% ahead of the NVIDIA GeForce GTX 980M, whose average score is 5308.

Average benchmark results indicate a meaningful performance gap: the GTX 760M averages 5236 across all tests, while the R5 M320 averages 4657: a difference of 11.1% when comparing their recorded average scores. The GTX 760M's 10.9% OpenCL lead and 14.2% Vulkan win translate into a consistent pattern: NVIDIA leads every test, and the margin is larger in modern compute workloads than in legacy ones.

Architecture Differences

The two GPUs come from different architectural generations and design philosophies. The NVIDIA GeForce GTX 760M uses the GK106S chip, built on Kepler architecture, built on a 28 nm process node at TSMC foundry. It packs 2,540 million transistors into a 221 mm² die, giving a transistor density of 11.5M per mm². The AMD Radeon R5 M320 uses the Jet chip, built on GCN 1.0 architecture, also on a 28 nm node at TSMC. Its die is far smaller at 56 mm², containing 690 million transistors, with a slightly higher transistor density of 12.3M per mm².

The shading resources reflect this scale difference: the GTX 760M has 768 shading units, 64 TMUs, and 16 ROPs; the R5 M320 has 320 shading units, 20 TMUs, and 8 ROPs. Pixel throughput reflects ROP count: the GTX 760M hits 11.50 GPixel/s, while the R5 M320 manages 6.84 GPixel/s. Texture rate is 46.02 GTexel/s for NVIDIA and 17.10 GTexel/s for AMD. FP32 compute output is 1,104.4 GFLOPS for the GTX 760M and 547.2 GFLOPS for the R5 M320, meaning NVIDIA delivers roughly double the raw FP32 throughput.

Memory design differs substantially. The GTX 760M uses 2 GB of GDDR5 on a 128 bit bus, with memory clocked at 1002 MHz or 4 Gbps effective, yielding 64.13 GB/s bandwidth. The R5 M320 pairs 4 GB of DDR3 on a 64 bit bus at 1000 MHz (2 Gbps effective), producing only 16.00 GB/s. The R5 M320 has more capacity, doubling the VRAM, but a quarter of the bandwidth. Clock behavior is also distinct: the GTX 760M has a base clock of 628 MHz and boost up to 719 MHz, while the R5 M320 has a higher base of 780 MHz and boost of 780 MHz and boost to 855 MHz. The GTX 760M draws 55 W TDP and uses an MXM module slot, while the R5 M320 lists no TDP and uses an IGP slot. The bus interface differs as well: PCIe 3.0 x16 for NVIDIA and PCIe 3.0 x8 for AMD.

API support is similar in scope, with Vulkan 1.2.175 on NVIDIA and 1.2.170 on AMD; DirectX support is 12 (11_0 for NVIDIA and 12 (11_1 for AMD, and OpenGL 4.6 on both.

The Verdict

From the recorded data, the NVIDIA GeForce GTX 760M is the stronger GPU for compute-heavy work: it wins both OpenCL and Vulkan tests, and leads by 10.1% in OpenCL and 14.2% in Vulkan in Vulkan. The R5 M320 offers more memory capacity, 4 GB versus 2 GB, and a higher boost clock, 855 MHz versus 719 MHz, but its performance scores are lower across every test. Its average score of 4657 is below the GTX 760M's 5236, by 11.1%. The GTX 760M's percentile ranking, 31st among all GPUs, also exceeds the R5 M320's 27th percentile. If the workload cares about raw compute, the GTX 760M is the choice. If the workload cares about VRAM capacity, the R5 M320 is the only one offering 4 GB, but the 64 bit bus and DDR3 memory type severely limit what that capacity can do. The data suggests the GTX 760M is the more capable part for typical GPU-bound tasks, while the R5 M320 should only be considered where 4 GB of VRAM is an absolute requirement and performance is secondary.

Specification Differences

The two differ in almost every measured specification:

• Chip and architecture: NVIDIA GK106S on Kepler; AMD Jet on GCN 1.0
• Transistors: 2,540 million (NVIDIA) versus 690 million (AMD)
• Die size: 221 mm² (NVIDIA) versus 56 mm² (AMD)
• Transistor density: 11.5M / mm² (NVIDIA) versus 12.3M / mm² (AMD)
• Base clock: 628 MHz (NVIDIA) versus 780 MHz (AMD)
• Boost clock: 719 MHz (NVIDIA) versus 855 MHz (AMD)
• Memory size: 2 GB (NVIDIA) versus 4 GB (AMD)
• Memory type: GDDR5 (NVIDIA) versus DDR3 (AMD)
• Memory bus: 128 bit (NVIDIA) versus 64 bit (AMD)
• Memory clock: 1002 MHz / 4 Gbps effective (NVIDIA) versus 1000 MHz / 2 Gbps effective (AMD)
• Bandwidth: 64.13 GB/s (NVIDIA) versus 16.00 GB/s (AMD)
• Shading units: 768 (NVIDIA) versus 320 (AMD)
• TMUs: 64 (NVIDIA) versus 20 (AMD)
• ROPs: 16 (NVIDIA) versus 8 (AMD)
• Pixel rate: 11.50 GPixel/s (NVIDIA) versus 6.84 GPixel/s (AMD)
• Texture rate: 46.02 GTexel/s (NVIDIA) versus 17.10 GTexel/s (AMD)
• FP32: 1,104.4 GFLOPS (NVIDIA) versus 547.2 GFLOPS (AMD)
• TDP: 55 W (NVIDIA) versus not listed for AMD
• Slot width: MXM Module (NVIDIA) versus IGP (AMD)
• Bus interface: PCIe 3.0 x16 (NVIDIA) versus PCIe 3.0 x8 (AMD)
• DirectX: 12 (11_0) (NVIDIA) versus 12 (11_1) (AMD)
• Vulkan: 1.2.175 (NVIDIA) versus 1.2.170 (AMD)
• Release date: 2013-05-29 (NVIDIA) versus 2015-05-04 (AMD)
• Successor: GeForce 800M (NVIDIA) versus Polaris Mobile (AMD)
• Predecessor: GeForce 600M (NVIDIA) versus Solar System (AMD)

FAQ

Q: Which GPU has the higher benchmark score?

A: The NVIDIA GeForce GTX 760M. It scores 5604 in OpenCL and 4868 in Vulkan, compared to the R5 M320's 5051 and 4262. Its average score is 5236 versus 4657.

Q: Does the AMD Radeon R5 M320 have more memory?

A: Yes, it has 4 GB of DDR3 on a 64 bit bus, whereas the GTX 760M has 2 GB of GDDR5 on a 128 bit bus. However, the GTX 760M's bandwidth is 64.13 GB/s versus 16.00 GB/s.

Q: Which GPU supports a newer DirectX version?

A: The AMD Radeon R5 M320 supports DirectX 12 (11_1), while the NVIDIA GeForce GTX 760M supports DirectX 12 (11_0). Both support OpenGL 4.6.

Q: What is the transistor count difference?

A: The GTX 760M has 2,540 million transistors on a 221 mm² die; the R5 M320 has 690 million on a 56 mm² die. The R5 M320 has a higher transistor density at 12.3M per mm² versus 11.5M per mm².

Q: Which GPU has a higher boost clock?

A: The AMD Radeon R5 M320 boosts to 855 MHz, higher than the GTX 760M's 719 MHz. The R5 M320 also has a higher base clock of 780 MHz versus 628 MHz.

Q: How do their Vulkan scores compare?

A: The GTX 760M scores 4868 in Vulkan, which is 14.2% higher than the R5 M320's 4262. In OpenCL, the GTX 760M leads by 10.9% (5604 versus 5051).

Where Each One Wins

NVIDIA GeForce GTX 760M wins on: raw compute performance across both tested workloads, with a 10.9% lead in OpenCL and 14.2% in Vulkan, a higher average score of 5236 versus 4657, a higher percentile rank at 31 versus 27, double the memory bandwidth at 64.13 GB/s versus 16.00 GB/s, more shading units at 768 versus 320, more TMUs at 64 versus 20, more ROPs at 16 versus 8, and higher pixel and texture rates at 11.50 GPixel/s and 46.02 GTexel/s versus 6.84 GPixel/s and 17.10 GTexel/s.

AMD Radeon R5 M320 wins on: memory capacity with 4 GB versus 2 GB, higher base and boost clocks at 780/855 MHz versus 628/719 MHz, a newer DirectX feature level at 12 (11_1 versus 12 (11_0), a higher transistor density at 12.3M per mm² versus 11.5M, and a later release date of 2015-05-04 versus 2013-05-29.

For a practical decision: the GTX 760M is the part to pick for any workload that stresses the GPU's compute or memory throughput, since its FP32 output of 1,104.4 GFLOPS is double the R5 M320's 547.2 GFLOPS, and its memory bandwidth is four times higher. The R5 M320 makes sense only if the application requires 4 GB of VRAM and can tolerate a 64 bit DDR3 bus and a 16.00 GB/s ceiling. In every recorded benchmark, the GTX 760M comes out ahead, and its nearest rival comparisons show it holding its own against similarly positioned GPUs, while the R5 M320 sits closer to the bottom of the field.