AMD Radeon R9 M375X vs NVIDIA GeForce GTX 675MX Comparison

The NVIDIA GeForce GTX 675MX and AMD Radeon R9 M375X are both end-of-life mobile graphics processors from different architectural generations, targeting the same performance bracket. Benchmark data shows the GTX 675MX holds a decisive lead in compute workloads, while the R9 M375X counters with a more modern feature set and higher clock speeds. The overall average benchmark scores place them within 1.2% of each other, with the GTX 675MX scoring 8427 and the R9 M375X scoring 8325. This places both at the 43rd percentile of all GPUs, indicating they are firmly in the mid-range of mobile graphics performance for their respective eras.

FAQ

Q: Which GPU has the higher average benchmark score?

A: The NVIDIA GeForce GTX 675MX scores 8427 on average, which is 1.2% higher than the AMD Radeon R9 M375X’s average score of 8325. The deltaPct in the GTX 675MX’s nearestRivals list confirms the R9 M375X is 1.2% behind.

Q: How do the two GPUs compare in OpenCL performance?

A: In the head-to-head Geekbench OpenCL test, the GTX 675MX scores 10723 compared to the R9 M375X’s 8273. This represents a 29.6% advantage for the NVIDIA part, making it the clear winner in this specific compute benchmark.

Q: What are the architectural differences between these two chips?

A: The GTX 675MX is built on NVIDIA’s Kepler architecture using the GK104 chip, while the R9 M375X uses AMD’s GCN 1.0 architecture with the Tropo chip. Both are manufactured on a 28 nm process at TSMC, but the Kepler chip has 3,540 million transistors on a 294 mm² die, whereas the GCN chip has 1,500 million transistors on a 123 mm² die.

Q: Which GPU supports a higher DirectX version?

A: The AMD Radeon R9 M375X supports DirectX 12 (11_1), while the NVIDIA GeForce GTX 675MX supports DirectX 12 (11_0). The AMD part has a slightly higher feature level in terms of DirectX support.

Q: What is the memory bandwidth difference between the two?

A: The GTX 675MX has a 256-bit memory bus providing 115.2 GB/s of bandwidth, while the R9 M375X has a 128-bit bus providing 72.00 GB/s. The NVIDIA GPU offers significantly higher memory bandwidth due to its wider bus interface.

Q: How do the pixel and texture rates compare?

A: The R9 M375X achieves a higher pixel rate of 16.24 GPixel/s versus 13.08 GPixel/s for the GTX 675MX. However, the GTX 675MX has a higher texture rate of 52.32 GTexel/s compared to 40.60 GTexel/s for the AMD part.

Architecture Differences

The two GPUs represent fundamentally different design philosophies from their respective manufacturers. The NVIDIA GeForce GTX 675MX is built on the Kepler architecture, utilizing the GK104 chip with 960 shading units, 80 texture mapping units, and 32 render output units. In contrast, the AMD Radeon R9 M375X uses the GCN 1.0 architecture with the Tropo chip, featuring 640 shading units, 40 TMUs, and only 16 ROPs. This means the NVIDIA GPU has 50% more shading units, double the TMUs, and double the ROPs compared to the AMD part.

The transistor counts reveal a stark difference in die complexity. The GK104 chip packs 3,540 million transistors into a 294 mm² die, while the Tropo chip contains 1,500 million transistors on a much smaller 123 mm² die. Interestingly, the transistor density is nearly identical at 12.0M / mm² for the NVIDIA chip and 12.2M / mm² for the AMD chip, indicating both were produced on the same 28 nm TSMC process with similar design efficiency. The larger die size for the GTX 675MX allows for more functional units, which explains its advantage in shading, texturing, and ROP throughput.

Clock speeds tell a different story. The R9 M375X runs at a base clock of 925 MHz with a boost clock of 1015 MHz, whereas the GTX 675MX has no listed base or boost clocks in the data. The AMD part’s higher operational frequency partially compensates for its smaller configuration, particularly in pixel rate where it achieves 16.24 GPixel/s compared to the GTX 675MX’s 13.08 GPixel/s. However, the GTX 675MX still dominates in texture rate at 52.32 GTexel/s versus 40.60 GTexel/s, thanks to its larger number of TMUs.

Memory subsystems also differ substantially. Both use 2 GB of GDDR5 memory, but the GTX 675MX employs a 256-bit bus delivering 115.2 GB/s of bandwidth, while the R9 M375X uses a 128-bit bus with 72.00 GB/s. The memory clock for the NVIDIA part is 900 MHz (3.6 Gbps effective), while the AMD part runs at 1125 MHz (4.5 Gbps effective). The R9 M375X’s faster memory clock cannot overcome the GTX 675MX’s doubled bus width, resulting in a substantial bandwidth deficit.

The interface and power characteristics also differ. The GTX 675MX uses an MXM-B (3.0) bus interface with an MXM Module slot width and no power connectors, rated at 100 W TDP. The R9 M375X uses a standard PCIe 3.0 x16 interface with no TDP specified in the data. The NVIDIA part is from the GeForce 600M generation released in 2012, while the AMD part belongs to the Gem System (R9 M300) generation released in 2015, a gap of roughly three years in release timing.

Head-to-Head Benchmarks

The only direct head-to-head benchmark available is Geekbench OpenCL, and the results are unequivocal. The NVIDIA GeForce GTX 675MX scores 10723 points compared to the AMD Radeon R9 M375X’s 8273 points. This yields a 29.6% delta in favor of the NVIDIA GPU. This substantial margin highlights the Kepler architecture’s strength in compute workloads, likely driven by the GTX 675MX’s higher shading unit count and memory bandwidth.

Examining the broader benchmark landscape, the GTX 675MX also has a Geekbench Metal score of 6131, though the R9 M375X does not have a corresponding Metal score in the data. However, the R9 M375X does have a Geekbench Vulkan score of 8377, which the GTX 675MX lacks. This divergence in API support reflects the architectural differences, with the AMD part offering a more modern Vulkan implementation (version 1.2.170) compared to the NVIDIA part’s Vulkan 1.2.175, though both support similar API versions.

The average benchmark scores place the GTX 675MX at 8427 and the R9 M375X at 8325. This narrow 102-point gap (1.2%) is interesting given the 29.6% delta in the OpenCL test. The discrepancy suggests that the newer AMD part performs relatively better in other workload types not captured in the head-to-head comparison, likely benefiting from its higher clock speeds and more efficient architecture per-transistor. The nearestRivals data confirms this close relationship: the R9 M375X lists the GTX 675MX as its closest rival with a -1.2% delta, while the GTX 675MX lists the R9 M375X with a +1.2% delta.

Both GPUs are surrounded by similar-performance competitors in their respective ranks. The GTX 675MX’s nearest rivals include the AMD Radeon 880M (8436, -0.1%), NVIDIA GeForce MX330 (8458, -0.4%), and AMD Radeon HD 8870M (8462, -0.4%). The R9 M375X’s nearest rivals include the NVIDIA Quadro K1200 (8265, +0.7%), AMD Radeon 880M (8436, -1.3%), and NVIDIA GeForce MX330 (8458, -1.6%). This clustering indicates that both GPUs sit in a tightly contested performance tier where architectural efficiency and driver optimization often determine the outcome.

Specification Differences

The two GPUs differ across nearly every major specification category. The most striking differences are in core configuration: the GTX 675MX has 960 shading units, 80 TMUs, and 32 ROPs, while the R9 M375X has 640 shading units, 40 TMUs, and 16 ROPs. This represents a 50% advantage in shading units and a 100% advantage in both TMUs and ROPs for the NVIDIA part.

Clock specifications show the R9 M375X with a base clock of 925 MHz and boost clock of 1015 MHz, while the GTX 675MX has no base or boost clocks listed. The memory clocks also differ: the GTX 675MX runs at 900 MHz (3.6 Gbps effective), while the R9 M375X runs at 1125 MHz (4.5 Gbps effective). The memory bus width is 256-bit for the NVIDIA part versus 128-bit for the AMD part, leading to bandwidth figures of 115.2 GB/s and 72.00 GB/s respectively.

Performance rates highlight the trade-offs between the two designs. The GTX 675MX achieves 13.08 GPixel/s pixel rate and 52.32 GTexel/s texture rate, while the R9 M375X achieves 16.24 GPixel/s and 40.60 GTexel/s respectively. For floating-point compute, the GTX 675MX delivers 1,255.7 GFLOPS FP32, while the R9 M375X delivers 1,299.2 GFLOPS FP32, giving the AMD part a slight edge in raw FP32 throughput.

The transistor and die specifications show the GTX 675MX with 3,540 million transistors on a 294 mm² die, versus 1,500 million transistors on a 123 mm² die for the R9 M375X. The bus interface differs as well: MXM-B (3.0) for the GTX 675MX versus PCIe 3.0 x16 for the R9 M375X. The TDP is listed at 100 W for the GTX 675MX, with no TDP specified for the R9 M375X. The NVIDIA part has a slot width of MXM Module and no power connectors, while the AMD part has no slot width or power connector data.

Both GPUs support 2 GB of GDDR5 memory, and both offer OpenGL 4.6. The DirectX support differs slightly, with the GTX 675MX supporting 12 (11_0) and the R9 M375X supporting 12 (11_1). Vulkan support is nearly identical at 1.2.175 for NVIDIA and 1.2.170 for AMD. The display outputs are listed as "Portable Device Dependent" for the GTX 675MX, while no display output data is available for the R9 M375X.

Where Each One Wins

The NVIDIA GeForce GTX 675MX wins decisively in compute-heavy workloads, as demonstrated by its 29.6% advantage in the Geekbench OpenCL benchmark. Its higher shading unit count (960 vs 640), doubled TMU and ROP counts, and substantially higher memory bandwidth (115.2 GB/s vs 72.00 GB/s) make it the stronger choice for applications that leverage parallel compute, such as GPGPU tasks, video encoding, and certain scientific workloads. The GTX 675MX also holds an advantage in texture-heavy scenarios, with its 52.32 GTexel/s texture rate outpacing the R9 M375X’s 40.60 GTexel/s, suggesting better performance in games and applications with complex texture filtering requirements.

The AMD Radeon R9 M375X wins in scenarios where its architectural advantages come to the fore. Its higher clock speeds (925 MHz base, 1015 MHz boost) and superior FP32 compute throughput (1,299.2 GFLOPS vs 1,255.7 GFLOPS) give it an edge in compute tasks that are less dependent on memory bandwidth and more on raw arithmetic throughput. The R9 M375X also achieves a higher pixel rate of 16.24 GPixel/s versus 13.08 GPixel/s for the GTX 675MX, indicating better fill-rate performance in pixel-heavy operations. Its newer release date (2015 vs 2012) and DirectX 12 (11_1) support suggest potentially better compatibility with modern API features, despite the close overall benchmark scores.

The two GPUs are closely matched in overall performance, with only 1.2% separating their average benchmark scores. The GTX 675MX’s victory in the head-to-head OpenCL test (10723 vs 8273) is the single largest margin in the data, but the R9 M375X’s Vulkan score of 8377, which has no direct NVIDIA counterpart in the data, hints at competitive performance in Vulkan-based workloads. For users prioritizing memory bandwidth, texture throughput, and compute performance, the GTX 675MX is the clear choice. For those valuing clock speed efficiency, raw FP32 throughput, and a more recent architectural foundation, the R9 M375X offers its own strengths. Given the narrow overall performance gap and the different workload characteristics, the choice between these two depends heavily on the specific applications and API requirements of the target use case.