AMD Radeon R7 M370 vs NVIDIA GeForce GTX 460 Comparison

# Head-to-Head Benchmarks

The only shared benchmark in the database is Geekbench OpenCL, and the results show a clear edge for the NVIDIA GeForce GTX 460. The GTX 460 scores 7,925 points, while the AMD Radeon R7 M370 manages 7,063 points. This translates to a delta of 12.2 percent in favor of the NVIDIA part. It is a decisive victory, not a narrow one, and it sets the tone for the rest of this comparison.

Looking at the broader context, the GTX 460 sits at the 41st percentile of all GPUs in the database, while the R7 M370 sits at the 38th percentile. The gap in raw performance is consistent with the percentile difference, though the delta is larger than the percentile spread might suggest. The GTX 460's average benchmark score is 7,925, while the R7 M370's average is 6,764, a difference of roughly 1,161 points when accounting for the R7 M370's additional Vulkan score of 6,465 dragging its average down.

The nearest rivals for the GTX 460 are instructive. The NVIDIA Quadro K4100M scores 7,906, just 0.2 percent behind the GTX 460, which places the two effectively at parity. The NVIDIA Quadro P5000 scores 8,039, 1.4 percent ahead, and the NVIDIA GeForce GTX 880M scores 8,040, also 1.4 percent ahead. The NVIDIA GeForce GTX 650 Ti scores 8,053, 1.6 percent ahead. The GTX 460 is therefore at the lower edge of a tight cluster, but it is competitive with much newer professional and mobile parts.

For the R7 M370, the nearest rivals paint a similar picture of parity with its peers. The AMD FirePro M5100 scores 6,830, 1 percent behind the R7 M370. The NVIDIA GeForce GT 1010 scores 6,698, 1 percent ahead of the R7 M370. The AMD Radeon R7 M460 scores 6,612, 2.3 percent ahead, and the NVIDIA GeForce GTX 675M scores 6,946, 2.6 percent behind. The R7 M370 is right in the middle of its own cluster, but that cluster sits roughly 1,000 points lower than the GTX 460's cluster.

The head-to-head benchmark count is one win for the GTX 460 and zero for the R7 M370. There is no Vulkan result for the GTX 460 in the database, so the R7 M370's Vulkan score of 6,465 cannot be directly compared. The OpenCL result is the only apples-to-apples measurement, and it favors the older NVIDIA card by a meaningful margin.

# Architecture Differences

The two cards come from different architectural generations and are built on different process nodes. The GTX 460 uses the GF104 chip, based on the Fermi architecture, fabricated at 40 nm by TSMC. The die size is 332 mm² and it packs 1,950 million transistors, which yields a transistor density of 5.9 million per square millimeter. This is a large, power-hungry chip by modern standards, but it was designed for desktop performance in 2010.

The R7 M370 uses the Litho chip, based on GCN 1.0 architecture, fabricated at 28 nm, also by TSMC. The die size is just 77 mm², with 950 million transistors, giving a transistor density of 12.3 million per square millimeter. The R7 M370 is a much smaller, denser chip, more than twice as dense as the GTX 460. This density advantage is a direct result of the more modern 28 nm process versus the older 40 nm node.

The shading units differ as well. The GTX 460 has 336 shading units, 56 texture mapping units, and 24 raster output pipelines. The R7 M370 has 384 shading units, 24 texture mapping units, and only 8 raster output pipelines. Despite having more shading units, the R7 M370 has far fewer TMUs and ROPs, which explains its lower texture and pixel rates. The GTX 460 delivers 37.80 GTexel/s and 9.450 GPixel/s, while the R7 M370 delivers 23.04 GTexel/s and 7.680 GPixel/s.

The FP32 compute throughput also favors the GTX 460. It delivers 907.2 GFLOPS, while the R7 M370 delivers 737.3 GFLOPS. The GTX 460 is ahead by roughly 23 percent in raw floating-point performance, which aligns with its OpenCL benchmark lead.

Memory configurations are notably different. The GTX 460 has 768 MB of GDDR5 on a 192-bit bus, yielding 86.40 GB/s of bandwidth. The R7 M370 has 2 GB of GDDR5 on a 128-bit bus, yielding 57.60 GB/s. The GTX 460 has higher bandwidth but far less capacity. The R7 M370's memory clock is the same effective 3.6 Gbps, but the narrower bus limits its throughput.

API support differs. The GTX 460 supports DirectX 12 (11_0) and OpenGL 4.6, with no Vulkan support listed. The R7 M370 supports DirectX 12 (11_1), OpenGL 4.6, and Vulkan 1.2.170. The R7 M370 has a more modern feature set, particularly Vulkan, which the GTX 460 lacks entirely.

# Where Each One Wins

The GTX 460 wins in raw compute performance, as measured by the OpenCL benchmark. It also wins in texture throughput, pixel throughput, and memory bandwidth. For workloads that are heavy on texture sampling, pixel fill, or memory bandwidth, the GTX 460 has a clear advantage. Its 56 TMUs versus 24 on the R7 M370 means it can process far more texture operations per second, and its 86.40 GB/s of bandwidth is nearly 50 percent higher than the R7 M370's 57.60 GB/s.

The R7 M370 wins on memory capacity, offering 2 GB versus 768 MB. For workloads that require large framebuffers or datasets that exceed 768 MB, the R7 M370 avoids the capacity wall that the GTX 460 would hit. The R7 M370 also wins on API modernity, with Vulkan support, which the GTX 460 lacks. Any application that uses Vulkan exclusively would not run on the GTX 460 at all.

The R7 M370 also wins on power efficiency, although the database does not list a TDP for it. The GTX 460 has a TDP of 160 W and requires dual-slot cooling with two 6-pin power connectors and a 450 W suggested power supply. The R7 M370 has no listed TDP, power connectors, or slot width, which suggests it is designed for mobile or low-power environments where such specifications are not relevant.

In terms of pure compute density, the R7 M370's 28 nm process gives it a transistor density of 12.3M per square millimeter, more than double the GTX 460's 5.9M. This is a manufacturing advantage, not a performance advantage, but it explains why the R7 M370 can exist in a much smaller footprint.

# Specification Differences

The two cards differ in nearly every specification category. The GTX 460 is built on 40 nm, while the R7 M370 is on 28 nm. The GTX 460 has 1,950 million transistors on a 332 mm² die, while the R7 M370 has 950 million on 77 mm². Transistor density is 5.9M versus 12.3M per square millimeter.

The GTX 460 has a memory clock of 900 MHz (3.6 Gbps effective), which is identical to the R7 M370's memory clock. However, the memory size differs: 768 MB versus 2 GB. The bus width is 192-bit versus 128-bit, and bandwidth is 86.40 GB/s versus 57.60 GB/s.

Shading units: 336 versus 384. TMUs: 56 versus 24. ROPs: 24 versus 8. Pixel rate: 9.450 GPixel/s versus 7.680 GPixel/s. Texture rate: 37.80 GTexel/s versus 23.04 GTexel/s. FP32: 907.2 GFLOPS versus 737.3 GFLOPS.

The GTX 460 has a base clock that is not listed, while the R7 M370 has a base clock of 875 MHz and a boost clock of 960 MHz. The GTX 460 has no boost clock listed.

The GTX 460 has a TDP of 160 W, a dual-slot form factor, two 6-pin power connectors, and a suggested PSU of 450 W. The R7 M370 has none of these listed, indicating it is a lower-power or mobile part.

The GTX 460 uses PCIe 2.0 x16, while the R7 M370 uses PCIe 3.0 x8. The GTX 460 has display outputs of 2x DVI and 1x mini-HDMI 1.3a; the R7 M370 has no display outputs listed.

The GTX 460 supports DirectX 12 (11_0), OpenGL 4.6, and no Vulkan. The R7 M370 supports DirectX 12 (11_1), OpenGL 4.6, and Vulkan 1.2.170.

The GTX 460 was released on 2010-07-11, while the R7 M370 was released on 2015-05-04. The GTX 460 has a launch MSRP of 199 USD; the R7 M370 has no launch MSRP listed.

# FAQ

Q: Which GPU is faster in the Geekbench OpenCL test?

A: The NVIDIA GeForce GTX 460 scores 7,925, which is 12.2 percent ahead of the AMD Radeon R7 M370's 7,063.

Q: Does the GTX 460 have more memory bandwidth?

A: Yes. The GTX 460 has 86.40 GB/s of bandwidth on a 192-bit bus, while the R7 M370 has 57.60 GB/s on a 128-bit bus.

Q: Which card supports Vulkan?

A: Only the AMD Radeon R7 M370 supports Vulkan, with version 1.2.170. The NVIDIA GeForce GTX 460 has no Vulkan support listed.

Q: What is the memory capacity difference?

A: The GTX 460 has 768 MB of GDDR5, while the R7 M370 has 2 GB of GDDR5. The R7 M370 has over 2.5 times the capacity.

Q: How does the FP32 compute performance compare?

A: The GTX 460 delivers 907.2 GFLOPS, which is higher than the R7 M370's 737.3 GFLOPS.

Q: Which card has a higher pixel fill rate?

A: The GTX 460 has a pixel rate of 9.450 GPixel/s, compared to the R7 M370's 7.680 GPixel/s.

# The Verdict

The data points to a clear winner for raw performance: the NVIDIA GeForce GTX 460. It leads by 12.2 percent in the only shared benchmark, has higher texture and pixel rates, higher memory bandwidth, and higher FP32 compute. For any workload that relies on these metrics, the GTX 460 is the stronger choice.

However, the AMD Radeon R7 M370 is not without merit. It offers 2 GB of memory versus 768 MB, which is a significant advantage for large datasets or high-resolution textures. It also supports Vulkan, which the GTX 460 cannot use. Its 28 nm process makes it a much smaller and likely more power-efficient part, though the database does not list a TDP for it.

The GTX 460 is a desktop card from 2010, with a 160 W TDP and dual-slot cooling. The R7 M370 is a mobile or low-profile part from 2015, with no listed power requirements. The GTX 460 will require a more substantial power supply and chassis, while the R7 M370 is designed to fit into constrained environments.

For users who prioritize raw compute, texture throughput, and memory bandwidth, the GTX 460 is the pick. For users who need more memory capacity, Vulkan support, or a smaller, lower-power footprint, the R7 M370 is the sensible choice. The benchmark data favors the GTX 460, but the R7 M370's feature set and capacity make it a viable alternative for specific use cases.