AMD Radeon R9 M360 vs NVIDIA GeForce GTX 675M Comparison

Head-to-Head Benchmarks

The head-to-head data contains a single recorded benchmark test: Geekbench OpenCL. In this test, the AMD Radeon R9 M360 scores 8211 points against the NVIDIA GeForce GTX 675M’s 6946 points. That is an 18.2% advantage for the AMD part. The delta is substantial enough to be considered a clear generational step rather than a marginal improvement.

To contextualize this, the database places the AMD Radeon R9 M360 at the 42nd percentile among all GPUs, with an average benchmark score of 8129 across its recorded tests. Its nearest rivals, according to the database, include the NVIDIA GeForce GTX 950M (average score 8135, a 0.1% difference) and the NVIDIA GeForce GTX 980 (average score 8167, a 0.5% difference). The AMD card sits within a tight cluster around the 8100 to 8170 mark, meaning its OpenCL score of 8211 is slightly above its own average and comfortably above the GTX 675M’s result.

The NVIDIA GeForce GTX 675M, by contrast, sits at the 39th percentile among all GPUs. Its average benchmark score is 6946, which is also its only recorded score. The database lists its nearest rivals as the AMD Radeon R5 M240 (average score 6975, a 0.4% difference) and the NVIDIA GeForce GTX 680M (average score 7023, a 1.1% difference). The GTX 675M’s result places it in a lower performance band, roughly 1200 to 1300 points below the AMD R9 M360 in raw OpenCL throughput.

The win count is 1 for the AMD Radeon R9 M360 and 0 for the NVIDIA GeForce GTX 675M in direct comparison. This is a single-test comparison, so the story is straightforward: in the one benchmark where both cards have recorded data, the AMD part leads by nearly a fifth. The absence of additional head-to-head tests (such as Vulkan or game-specific workloads) means the analysis rests on this one metric, but the margin is large enough to indicate a meaningful performance gap.

Architecture Differences

The two GPUs come from different architectural eras. The AMD Radeon R9 M360 uses the Tropo chip, built on GCN 1.0 architecture, and belongs to the Gem System (R9 M300) generation. It is fabricated on a 28 nm process at TSMC, with 1,500 million transistors on a 123 mm² die. That yields a transistor density of 12.2 million transistors per square millimeter. The NVIDIA GeForce GTX 675M, meanwhile, uses the GF114 chip, built on Fermi 2.0 architecture, and belongs to the GeForce 600M generation. It is fabricated on a 40 nm process at TSMC, with 1,950 million transistors on a 332 mm² die, giving a transistor density of 5.9 million transistors per square millimeter.

The process node difference is stark: 28 nm versus 40 nm. The AMD part packs more transistors into a much smaller die, which explains its higher density. The NVIDIA part uses an older, larger process, resulting in a die that is nearly three times larger physically despite having fewer transistors per area. This architectural gap likely contributes to the performance difference seen in the benchmark.

Both GPUs are listed as end-of-life products. The AMD R9 M360 was released on May 4, 2015, with its predecessor listed as the Solar System and its successor as the Polaris Mobile. The NVIDIA GTX 675M was released on March 21, 2012, with its predecessor listed as the GeForce 500M and its successor as the GeForce 700M. The AMD part is roughly three years newer, which aligns with its architectural advantages.

Feature support also differs. The AMD R9 M360 supports DirectX 12 (11_1), OpenGL 4.6, and Vulkan 1.2.170. The NVIDIA GTX 675M supports DirectX 12 (11_0) and OpenGL 4.6, but has no recorded Vulkan support. This means the AMD part is more future-proof for modern APIs, particularly Vulkan, which is increasingly used in cross-platform engines and emulators. The DirectX version gap, 11_1 versus 11_0, is minor but indicates the AMD part can handle slightly newer feature levels.

Where Each One Wins

The AMD Radeon R9 M360 wins the only benchmark where both cards have recorded data. Its OpenCL score of 8211 versus 6946 gives it an 18.2% lead. This is a compute-oriented workload, and the AMD part’s GCN architecture, with its 512 shading units, is well-suited to parallel compute tasks. Additionally, the AMD part has Vulkan support, which the NVIDIA GTX 675M lacks, so any Vulkan-based workload would be a win by default for the AMD card.

The NVIDIA GeForce GTX 675M has no recorded wins in this comparison. Its only benchmark score is the same 6946 OpenCL result that appears in the head-to-head. However, the database does not list any additional tests for this card, so it is possible that in certain workloads the GTX 675M could perform relatively better. The card has 64 texture mapping units and 32 ROPs, compared to the AMD part’s 32 TMUs and 16 ROPs, and a 256-bit memory bus versus the AMD part’s 128-bit bus. These specifications suggest the NVIDIA card may have advantages in texture-heavy or fill-rate-limited scenarios, even if the raw compute score is lower.

The AMD R9 M360 has a texture rate of 29.60 GTexel/s and a pixel rate of 14.80 GPixel/s. The NVIDIA GTX 675M has a texture rate of 39.68 GTexel/s and a pixel rate of 9.920 GPixel/s. So the NVIDIA part is faster at texturing, while the AMD part is faster at pixel filling. In modern games, texture rate often matters more than pixel rate, which could partially offset the AMD part’s compute lead in gaming workloads. However, without recorded gaming benchmarks, this remains an inference from the specification data.

Specification Differences

The two cards differ across nearly every major specification category. The AMD Radeon R9 M360 has 4 GB of GDDR5 memory on a 128-bit bus, with a bandwidth of 72.00 GB/s. The NVIDIA GeForce GTX 675M has 2 GB of GDDR5 memory on a 256-bit bus, with a bandwidth of 96.00 GB/s. The NVIDIA part has higher bandwidth due to its wider bus, but the AMD part has double the memory capacity.

Clock speeds also differ. The AMD R9 M360 has a base clock of 900 MHz and a boost clock of 925 MHz. The NVIDIA GTX 675M has no recorded base or boost clocks. For memory, the AMD part runs at 1125 MHz (4.5 Gbps effective), while the NVIDIA part runs at 750 MHz (3 Gbps effective). The AMD memory clock is higher, but the NVIDIA wider bus compensates.

Compute unit counts differ significantly. The AMD R9 M360 has 512 shading units, 32 TMUs, and 16 ROPs. The NVIDIA GTX 675M has 384 shading units, 64 TMUs, and 32 ROPs. The AMD part has more shading units, while the NVIDIA part has more TMUs and ROPs. The FP32 performance is close: 947.2 GFLOPS for the AMD part versus 952.3 GFLOPS for the NVIDIA part, a difference of less than 1%. This explains why the OpenCL scores are not as far apart as one might expect from the architecture gap.

Power and physical differences are notable. The NVIDIA GTX 675M has a TDP of 100 W, while the AMD R9 M360 has no recorded TDP. The NVIDIA card is an MXM Module with no power connectors, and its display outputs are listed as portable device dependent. The AMD card has a PCIe 3.0 x16 bus interface, while the NVIDIA card uses MXM-B (3.0). The AMD part has no recorded TDP, slot width, or power connector details.

FAQ

Q: Which GPU has the higher benchmark score?

A: The AMD Radeon R9 M360 scores 8211 in Geekbench OpenCL, while the NVIDIA GeForce GTX 675M scores 6946. That is an 18.2% difference in favor of the AMD part.

Q: How do these cards compare to their nearest rivals?

A: The AMD R9 M360 sits within 0.5% of the NVIDIA GeForce GTX 950M and the NVIDIA GeForce GTX 980, based on average scores. The NVIDIA GTX 675M sits within 1.3% of the NVIDIA GeForce GTX 680M and the NVIDIA T600.

Q: Does the NVIDIA GTX 675M have any specification advantages?

A: Yes. It has a 256-bit memory bus versus the AMD part’s 128-bit bus, yielding higher bandwidth at 96.00 GB/s versus 72.00 GB/s. It also has more TMUs (64 versus 32) and more ROPs (32 versus 16).

Q: Does the AMD R9 M360 support newer APIs?

A: Yes, the AMD part supports Vulkan 1.2.170, while the NVIDIA GTX 675M has no recorded Vulkan support. Both support DirectX 12, but the AMD part supports 11_1 while the NVIDIA part supports 11_0.

Q: Which card was released more recently?

A: The AMD Radeon R9 M360 was released on May 4, 2015. The NVIDIA GeForce GTX 675M was released on March 21, 2012. The AMD part is roughly three years newer.

Q: How do the memory capacities compare?

A: The AMD R9 M360 has 4 GB of GDDR5 memory. The NVIDIA GTX 675M has 2 GB of GDDR5 memory. Both use GDDR5, but the AMD part has double the capacity.

The Verdict

Based on the recorded data, the AMD Radeon R9 M360 is the faster GPU in compute workloads. Its 18.2% lead in Geekbench OpenCL is decisive, and its higher percentile ranking (42nd versus 39th) reinforces this. The AMD part also brings modern features like Vulkan support and a newer architecture (GCN 1.0 on 28 nm versus Fermi 2.0 on 40 nm). For users running OpenCL compute tasks or Vulkan-based applications, the AMD R9 M360 is the clear choice.

However, the NVIDIA GeForce GTX 675M is not without merit. Its higher bandwidth (96.00 GB/s versus 72.00 GB/s) and larger texture unit count (64 versus 32) suggest it may handle certain graphics workloads, particularly texture-heavy scenes, more efficiently. Its FP32 performance is nearly identical to the AMD part (952.3 GFLOPS versus 947.2 GFLOPS), so the gap in compute is not due to raw floating-point throughput but likely other architectural factors.

The database shows only one head-to-head benchmark, so conclusions about gaming performance are limited. The NVIDIA card has a 100 W TDP and is an MXM module, indicating it was designed for laptops. The AMD card has no recorded TDP, but its newer process node suggests it may be more power-efficient per unit of performance.

For a user prioritizing compute performance, modern API support, and memory capacity, the AMD Radeon R9 M360 is the stronger pick. For a user who values texture throughput and memory bandwidth, the NVIDIA GeForce GTX 675M offers specific advantages, though its older architecture and lack of Vulkan support are notable drawbacks. The data favors the AMD card overall, but the NVIDIA card’s specification strengths mean it could win in scenarios not covered by the recorded benchmarks.