AMD Radeon R9 M360 vs NVIDIA GeForce GTX 660 Comparison

Head-to-Head Benchmarks

The benchmark data reveals a clear and consistent winner across both shared test workloads. In the Geekbench OpenCL test, the NVIDIA GeForce GTX 660 scores 11,347, while the AMD Radeon R9 M360 trails at 8,211. That represents a 38.2% advantage for the NVIDIA card, a substantial margin that indicates a decisive performance gap in compute-heavy tasks. The Vulkan results tell a similar story, with the GTX 660 reaching 11,415 against the R9 M360's 8,047, an even larger 41.9% lead. Across the two head-to-head benchmarks, the GTX 660 claims both victories, giving it a 2-0 win record over the Radeon part.

Looking at the broader database context, the GTX 660's average benchmark score of 9,022 places it in the 45th percentile of all GPUs tracked. Its nearest rivals include the NVIDIA TITAN V CEO Edition at 9,037 (a negligible 0.2% gap), the GeForce GTX 560 at 9,058 (0.4% behind), the AMD Radeon 550X at 8,918 (1.2% ahead of the 550X), and the AMD Radeon Pro WX 5100 at 8,863 (1.8% ahead). These tight margins show the GTX 660 sits in a crowded performance tier, essentially matching far more modern or expensive parts in aggregate scoring.

The R9 M360, by contrast, averages 8,129 across its recorded benchmarks, placing it in the 42nd percentile. Its closest competitors include the NVIDIA GeForce GTX 950M at 8,135 (just 0.1% behind), the GeForce 945M at 8,099 (0.4% ahead), the GeForce GTX 980 at 8,167 (0.5% behind), and the NVIDIA GRID K2 at 8,080 (0.6% ahead). The R9 M360 sits squarely within a mobile-oriented performance band, with its average score nearly indistinguishable from several NVIDIA notebook parts.

When comparing the two directly, the GTX 660's OpenCL score is roughly 38% higher than the R9 M360's, and its Vulkan score is about 42% higher. These are not marginal differences; they represent a full performance class separation. The GTX 660's texture rate of 82.56 GTexel/s versus the R9 M360's 29.60 GTexel/s, along with a pixel rate of 20.64 GPixel/s versus 14.80 GPixel/s, corroborates the benchmark findings. The data indicates the GTX 660 delivers roughly 2.8 times the texture throughput and about 1.4 times the pixel throughput of the Radeon part.

In floating-point compute, the GTX 660 achieves 1.981 TFLOPS of FP32 performance, while the R9 M360 manages 947.2 GFLOPS. That is a 2.1x advantage for the NVIDIA card, which aligns closely with the observed benchmark deltas. The GTX 660's memory bandwidth of 144.2 GB/s over a 192-bit bus also dwarfs the R9 M360's 72.00 GB/s over a 128-bit interface, providing another structural explanation for the performance gap.

Architecture Differences

The two GPUs come from different architectural generations and design philosophies. The NVIDIA GeForce GTX 660 is built on the Kepler architecture, specifically the GK106 chip, fabricated on a 28 nm process at TSMC. It packs 2,540 million transistors onto a 221 mm² die, yielding a transistor density of 11.5 million per square millimeter. The AMD Radeon R9 M360 uses the GCN 1.0 architecture with the Tropo chip, also on a 28 nm TSMC process, but with 1,500 million transistors on a smaller 123 mm² die, giving it a slightly higher density of 12.2 million per square millimeter.

The execution resources differ dramatically. The GTX 660 fields 960 shading units, 80 texture mapping units, and 24 ROPs. The R9 M360, by contrast, has only 512 shading units, 32 TMUs, and 16 ROPs. This 1.875x advantage in shader count, 2.5x in TMUs, and 1.5x in ROPs for the NVIDIA card directly explains its superior fill rates and compute throughput. Clock speeds also favor NVIDIA: the GTX 660 runs at a base of 980 MHz with a boost of 1032 MHz, while the R9 M360 operates at 900 MHz base and 925 MHz boost.

Memory configurations reveal another fundamental split. The GTX 660 uses 2 GB of GDDR5 on a 192-bit bus, running at 1502 MHz with 6 Gbps effective data rate, producing 144.2 GB/s of bandwidth. The R9 M360 uses 4 GB of GDDR5 on a narrower 128-bit bus, at 1125 MHz with 4.5 Gbps effective, yielding just 72.00 GB/s. The Radeon card offers double the capacity but less than half the bandwidth, a tradeoff that favors the NVIDIA part in bandwidth-sensitive workloads.

API support shows minor differences. The GTX 660 supports DirectX 12 (11_0), OpenGL 4.6, and Vulkan 1.2.175. The R9 M360 supports DirectX 12 (11_1), OpenGL 4.6, and Vulkan 1.2.170. Both are end-of-life products, but the GTX 660 launched on September 5, 2012, while the R9 M360 arrived on May 4, 2015. The NVIDIA card is part of the GeForce 600 generation, succeeding the GeForce 500 series and preceding the GeForce 700 series. The AMD card belongs to the Gem System (R9 M300) generation, succeeding the Solar System and preceding Polaris Mobile.

The Verdict

The data points decisively to the NVIDIA GeForce GTX 660 as the stronger performer. It wins both shared benchmarks by margins of 38.2% and 41.9%, and its average benchmark score of 9,022 is about 11% higher than the R9 M360's 8,129. The GTX 660 also holds a higher percentile ranking at 45 versus 42, indicating better standing against the full GPU population in the database.

For users prioritizing raw compute performance, the GTX 660 is the clear choice. Its 1.981 TFLOPS of FP32 performance, 82.56 GTexel/s texture rate, and 144.2 GB/s memory bandwidth provide substantial headroom over the R9 M360's 947.2 GFLOPS, 29.60 GTexel/s, and 72.00 GB/s. The NVIDIA card's higher clock speeds and more generous execution resources make it suitable for tasks that stress shader throughput and memory bandwidth.

The R9 M360 does offer one practical advantage: 4 GB of memory versus 2 GB. For workloads that require larger framebuffers, such as high-resolution textures or multi-monitor setups, the additional capacity could be relevant. However, the significantly lower bandwidth means that capacity is accessed at a much slower rate, potentially negating the benefit in practice.

Given the recorded data, the GTX 660 is the superior GPU for compute-bound applications and gaming workloads that rely on shader performance. The R9 M360 might appeal to users who need more VRAM capacity and are willing to accept lower throughput. The choice ultimately hinges on whether memory size or memory speed matters more, and the benchmark results suggest that speed wins in the tests recorded.

Specification Differences

The two GPUs differ across nearly every major specification field:

• Architecture: Kepler (GK106) versus GCN 1.0 (Tropo)
• Transistors: 2,540 million versus 1,500 million
• Die Size: 221 mm² versus 123 mm²
• Transistor Density: 11.5M / mm² versus 12.2M / mm²
• Base Clock: 980 MHz versus 900 MHz
• Boost Clock: 1032 MHz versus 925 MHz
• Memory Clock: 1502 MHz (6 Gbps effective) versus 1125 MHz (4.5 Gbps effective)
• Memory Size: 2 GB versus 4 GB
• Memory Bus Width: 192 bit versus 128 bit
• Memory Bandwidth: 144.2 GB/s versus 72.00 GB/s
• Shading Units: 960 versus 512
• Texture Mapping Units: 80 versus 32
• Render Output Units: 24 versus 16
• Pixel Rate: 20.64 GPixel/s versus 14.80 GPixel/s
• Texture Rate: 82.56 GTexel/s versus 29.60 GTexel/s
• FP32 Performance: 1.981 TFLOPS versus 947.2 GFLOPS
• TDP: 140 W versus not specified
• API Support: DirectX 12 (11_0) versus DirectX 12 (11_1); Vulkan 1.2.175 versus 1.2.170
• Release Date: September 5, 2012 versus May 4, 2015
• Launch MSRP: 229 USD versus not available

Both use the same 28 nm TSMC process, PCIe 3.0 x16 interface, and GDDR5 memory type. The GTX 660 has a dual-slot form factor, a 1x 6-pin power connector, a 300 W suggested PSU, and dimensions of 241 mm (9.5 inches). The R9 M360 lacks recorded specifications for slot width, power connectors, suggested PSU, dimensions, and display outputs.

FAQ

Q: Which GPU wins in OpenCL performance?

A: The NVIDIA GeForce GTX 660 scores 11,347 in Geekbench OpenCL, which is 38.2% higher than the AMD Radeon R9 M360's 8,211.

Q: How do the two compare in Vulkan benchmarks?

A: The GTX 660 achieves 11,415, while the R9 M360 scores 8,047. The NVIDIA card leads by 41.9%.

Q: What is the memory capacity difference?

A: The R9 M360 has 4 GB of GDDR5, while the GTX 660 has 2 GB. However, the GTX 660 has 144.2 GB/s bandwidth versus 72.00 GB/s for the Radeon.

Q: Which GPU has more shading units?

A: The GTX 660 has 960 shading units, compared to 512 on the R9 M360.

Q: Are both GPUs the same process node?

A: Yes, both are fabricated on a 28 nm process at TSMC. The GTX 660 uses the GK106 chip, and the R9 M360 uses the Tropo chip.

Q: What are the average benchmark scores for each?

A: The GTX 660 averages 9,022 across its tests, while the R9 M360 averages 8,129. The GTX 660 sits at the 45th percentile of all GPUs, and the R9 M360 at the 42nd.