AMD Radeon R7 M370 vs NVIDIA GeForce GTX 650 Ti Comparison

Where Each One Wins

The benchmark data splits cleanly along workload type, with the NVIDIA GeForce GTX 650 Ti taking both recorded victories. In Geekbench OpenCL, the GTX 650 Ti scores 7,877 against the Radeon R7 M370’s 7,063, a margin of 11.5 percent. The gap widens considerably in Geekbench Vulkan, where the GTX 650 Ti posts 8,229 versus 6,465, an advantage of 27.3 percent. That Vulkan result is the single largest differentiator between these two parts, and it suggests the NVIDIA architecture handles compute-heavy, low-level API workloads with substantially more efficiency.

The Radeon R7 M370 does not win either recorded test, but its OpenCL showing is respectable in absolute terms. A score of 7,063 places it within striking distance of the GTX 650 Ti in that specific workload, and the delta of 11.5 percent is far narrower than the Vulkan gap. For users whose primary concern is OpenCL compute, the AMD part is not embarrassed. However, the Vulkan result paints a different picture: 6,465 is nearly 21.5 percent below the GTX 650 Ti’s own Vulkan score, and the R7 M370’s average benchmark score of 6,764 trails the GTX 650 Ti’s 8,053 by roughly 16 percent.

Looking at the percentile rankings, the GTX 650 Ti sits at the 42nd percentile of all GPUs in the database, while the R7 M370 lands at the 38th. That four-point spread is modest, but it aligns with the head-to-head results: the NVIDIA part is consistently ahead, though not by an overwhelming amount in every test. The GTX 650 Ti’s nearest rivals include the GeForce GTX 880M (delta 0.2 percent), the GTX 650 Ti Boost (delta -0.2 percent), and the Quadro P5000 (delta 0.2 percent), which puts it in a tightly clustered performance band. The R7 M370’s nearest rivals, by contrast, include the FirePro M5100 (delta -1 percent), the GeForce GT 1010 (delta 1 percent), and the Radeon R7 M460 (delta 2.3 percent), indicating it competes in a lower tier overall.

For use-case planning, the data indicates the GTX 650 Ti is the better choice for Vulkan-based applications and general compute throughput, while the R7 M370 can handle OpenCL tasks without collapsing, but it will lag in any scenario that stresses the GPU’s architectural efficiency. The R7 M370’s 2 GB framebuffer may help in memory-heavy scenarios, but the raw compute results show it cannot translate that capacity into higher scores.

Architecture Differences

The two GPUs come from different design philosophies and different eras. The GTX 650 Ti uses the GK106S chip built on NVIDIA’s Kepler architecture, manufactured by TSMC on a 28 nm process. It packs 2,540 million transistors into a 221 mm² die, yielding a transistor density of 11.5 million per square millimeter. The R7 M370 uses the Litho chip based on AMD’s GCN 1.0 architecture, also on TSMC’s 28 nm node, but with 950 million transistors on a 77 mm² die, for a density of 12.3 million per square millimeter. The density figures are close, but the sheer scale differs enormously: the NVIDIA chip has roughly 2.7 times more transistors and nearly three times the die area.

The compute resources reflect that disparity. The GTX 650 Ti carries 768 shading units, 64 texture mapping units, and 16 raster output units. The R7 M370 has 384 shading units, 24 TMUs, and 8 ROPs. In every category, the NVIDIA part has exactly double the shading units and TMUs, and double the ROPs. That structural advantage explains why the GTX 650 Ti’s peak pixel rate is 14.85 GPixel/s versus 7.68 GPixel/s for the AMD part, and its texture rate is 59.39 GTexel/s versus 23.04 GTexel/s. FP32 throughput tells a similar story: 1,425.4 GFLOPS for the GTX 650 Ti against 737.3 GFLOPS for the R7 M370.

Clock behavior differs as well. The R7 M370 has explicit base and boost clocks of 875 MHz and 960 MHz respectively, while the GTX 650 Ti’s base and boost clocks are not recorded in the database. Memory clocks are also distinct: the GTX 650 Ti runs at 1350 MHz with 5.4 Gbps effective, while the R7 M370 runs at 900 MHz with 3.6 Gbps effective. Both use GDDR5 on a 128-bit bus, but the GTX 650 Ti’s higher clock yields 86.40 GB/s of bandwidth versus 57.60 GB/s for the R7 M370. The AMD part does offer 2 GB of memory versus the GTX 650 Ti’s 1 GB, which is a real capacity advantage, but it comes with lower bandwidth.

The bus interfaces differ too. The GTX 650 Ti uses PCIe 3.0 x16, while the R7 M370 uses PCIe 3.0 x8. In practice, that means the NVIDIA part has twice the PCIe lanes available for data transfer, which can matter in compute workloads that stream data to and from system memory. API support is broadly similar: both support DirectX 12 (the GTX 650 Ti at 11_0 level, the R7 M370 at 11_1), OpenGL 4.6, and Vulkan (the GTX 650 Ti at 1.2.175, the R7 M370 at 1.2.170). The NVIDIA part is older, with a release date of October 8, 2012, versus May 4, 2015 for the AMD part, and the GTX 650 Ti carried a launch MSRP of 149 USD, while the R7 M370 has no recorded MSRP.

FAQ

Q: Which GPU is faster in Vulkan workloads?

A: The NVIDIA GeForce GTX 650 Ti. It scores 8,229 in Geekbench Vulkan, which is 27.3 percent ahead of the Radeon R7 M370’s 6,465.

Q: Does the Radeon R7 M370 win any benchmark in the database?

A: No. The head-to-head results show the GTX 650 Ti winning both recorded tests, with 2 wins and 0 losses for the AMD part.

Q: How much memory does each GPU have?

A: The Radeon R7 M370 has 2 GB of GDDR5, while the GeForce GTX 650 Ti has 1 GB of GDDR5. Both use a 128-bit memory bus.

Q: What is the transistor count difference?

A: The GTX 650 Ti has 2,540 million transistors on a 221 mm² die, while the R7 M370 has 950 million on a 77 mm² die.

Q: Which GPU has higher memory bandwidth?

A: The GeForce GTX 650 Ti, with 86.40 GB/s versus 57.60 GB/s for the Radeon R7 M370, despite the AMD card having twice the memory capacity.

Q: When was each GPU released?

A: The GTX 650 Ti was released on October 8, 2012, and the R7 M370 was released on May 4, 2015. Both are now end-of-life products.

Specification Differences

The two GPUs differ across nearly every measurable specification. The GTX 650 Ti uses the GK106S chip with Kepler architecture, while the R7 M370 uses the Litho chip with GCN 1.0 architecture. Both are built on TSMC’s 28 nm process, but the GTX 650 Ti has 2,540 million transistors versus 950 million, and a die size of 221 mm² versus 77 mm². Transistor density is slightly higher on the AMD part at 12.3M per mm², compared to 11.5M per mm² for NVIDIA.

The shading units are 768 on the GTX 650 Ti and 384 on the R7 M370. TMUs are 64 versus 24, and ROPs are 16 versus 8. Pixel rate is 14.85 GPixel/s versus 7.68 GPixel/s, and texture rate is 59.39 GTexel/s versus 23.04 GTexel/s. FP32 performance is 1,425.4 GFLOPS against 737.3 GFLOPS. The GTX 650 Ti has no recorded base or boost clock, while the R7 M370 has a base clock of 875 MHz and a boost clock of 960 MHz.

Memory configuration diverges on capacity and speed. The GTX 650 Ti offers 1024 MB of GDDR5 at 1350 MHz (5.4 Gbps effective), yielding 86.40 GB/s bandwidth. The R7 M370 offers 2048 MB of GDDR5 at 900 MHz (3.6 Gbps effective), yielding 57.60 GB/s bandwidth. Both use a 128-bit bus. The GTX 650 Ti uses PCIe 3.0 x16, while the R7 M370 uses PCIe 3.0 x8. The GTX 650 Ti has a TDP of 110 W, a single-slot form factor, and a 1x 6-pin power connector, with a suggested PSU of 300 W; none of those figures are recorded for the R7 M370. The GTX 650 Ti’s display outputs are 2x DVI and 1x mini-HDMI 1.4a, while the R7 M370 has no recorded display outputs. API support is similar but not identical: DirectX 12 (11_0) for the GTX 650 Ti versus DirectX 12 (11_1) for the R7 M370, OpenGL 4.6 for both, and Vulkan 1.2.175 versus 1.2.170.

Head-to-Head Benchmarks

The Geekbench OpenCL test favors the GTX 650 Ti, but the margin is moderate. NVIDIA’s card scores 7,877, while the R7 M370 manages 7,063, putting the GTX 650 Ti 11.5 percent ahead. That result is consistent with the raw compute specifications: the GTX 650 Ti has double the shading units and more than double the texture rate, so a double-digit lead in a compute-heavy API is expected. Yet the R7 M370’s score is not catastrophic, and it sits within 12 percent of the NVIDIA part despite having half the shader count. This suggests the GCN architecture extracts more efficiency per shading unit in OpenCL, or the higher boost clock on the AMD part helps close the gap.

The Geekbench Vulkan test tells a very different story. The GTX 650 Ti leaps to 8,229, while the R7 M370 drops to 6,465. The delta is 27.3 percent, more than double the OpenCL margin. Vulkan’s low-level access to the hardware appears to amplify the architectural advantages of the Kepler design, or it exposes weaknesses in the GCN 1.0 implementation on the R7 M370. Notably, the GTX 650 Ti’s Vulkan score is higher than its OpenCL score by about 4.5 percent, while the R7 M370’s Vulkan score is lower than its OpenCL score by about 8.5 percent. That divergence is the clearest signal in the data: the NVIDIA part benefits from Vulkan, while the AMD part loses ground.

The average benchmark scores reinforce the overall hierarchy. The GTX 650 Ti averages 8,053, and its nearest rivals are all within a narrow band: the GTX 880M at 8,040 (0.2 percent delta), the GTX 650 Ti Boost at 8,067 (-0.2 percent), the Quadro P5000 at 8,039 (0.2 percent), and the GRID K2 at 8,080 (-0.3 percent). This places the GTX 650 Ti in a very tight competitive cluster where small differences decide rankings. The R7 M370 averages 6,764, and its nearest rivals are the FirePro M5100 at 6,830 (-1 percent), the GeForce GT 1010 at 6,698 (1 percent), the Radeon R7 M460 at 6,612 (2.3 percent), and the GeForce GTX 675M at 6,946 (-2.6 percent). The spreads are wider here, indicating less consistency among the AMD part’s competitors.

The largest single win for the GTX 650 Ti is the 27.3 percent Vulkan margin. That result alone accounts for most of the average score difference between the two cards. If the R7 M370 had matched its OpenCL performance in Vulkan, the overall comparison would be much closer, but it did not. The data shows a GPU that is competitive in one API and significantly behind in another. For any workload that relies on Vulkan, the GTX 650 Ti is the clear choice. For OpenCL-only scenarios, the R7 M370 is within a reasonable margin, but it still loses the head-to-head test. The verdict is unambiguous: the GTX 650 Ti wins both recorded benchmarks and holds a higher percentile ranking, making it the stronger GPU in this pairing.