AMD Radeon R7 M370 vs NVIDIA GeForce GTX 650 Ti Boost Comparison

Where Each One Wins

The benchmark data splits cleanly in favor of one side. The NVIDIA GeForce GTX 650 Ti Boost wins both recorded head-to-head tests, and the margins are substantial. In Geekbench OpenCL, the NVIDIA card scores 9302 against 7063 for the AMD Radeon R7 M370, a 31.7% advantage. In Geekbench Vulkan, the gap widens to 55.3%, with scores of 10041 and 6465 respectively. There is no test in the database where the AMD part finishes ahead.

This is not a close contest. The NVIDIA card's average benchmark score sits at 8067, placing it in the 42nd percentile of all GPUs, while the AMD card averages 6764, good for the 38th percentile. The NVIDIA card also benefits from having a third benchmark result, Geekbench Metal at 4858, which the AMD card lacks entirely. That additional data point reinforces the NVIDIA card's position as the more thoroughly tested and consistently faster option.

The use-case split, therefore, is not about which card wins which workload. It is about what each card can credibly handle given its performance envelope. The GTX 650 Ti Boost is the choice for any task that stresses compute throughput, whether OpenCL or Vulkan. The R7 M370 remains a functional option for lighter duties, but the recorded data gives it no victory condition.

FAQ

Q: Which GPU has the higher average benchmark score?

A: The NVIDIA GeForce GTX 650 Ti Boost averages 8067 across its benchmark results, compared to 6764 for the AMD Radeon R7 M370.

Q: How large is the performance gap in OpenCL compute?

A: The GTX 650 Ti Boost scores 9302 in Geekbench OpenCL, which is 31.7% higher than the R7 M370's 7063.

Q: Does the AMD Radeon R7 M370 win any head-to-head benchmark?

A: No. The database records two head-to-head tests, OpenCL and Vulkan, and the NVIDIA card wins both.

Q: What is the Vulkan performance difference?

A: The GTX 650 Ti Boost scores 10041 in Geekbench Vulkan, while the R7 M370 scores 6465, a 55.3% advantage for NVIDIA.

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

A: The GTX 650 Ti Boost sits within 0.3% of the NVIDIA GRID K2, GTX 650 Ti, GTX 880M, and Quadro P5000. The R7 M370 is within 2.6% of the AMD FirePro M5100, GT 1010, R7 M460, and GTX 675M.

Q: Which GPU has more shading units?

A: The NVIDIA card has 768 shading units, exactly double the 384 found in the AMD Radeon R7 M370.

Head-to-Head Benchmarks

The head-to-head results tell a story of consistent NVIDIA dominance. In Geekbench OpenCL, the GTX 650 Ti Boost posts 9302 points against the R7 M370's 7063. That 31.7% delta is significant, but it is not the largest gap in this comparison. The Vulkan test shows an even more pronounced separation: 10041 for NVIDIA versus 6465 for AMD, a 55.3% lead.

Why does Vulkan favor the NVIDIA card so much more than OpenCL? The architecture differences explain part of it. The GK106 chip in the GTX 650 Ti Boost carries 768 shading units, 64 texture mapping units, and 24 ROPs. The AMD Litho chip has half the shading units at 384, 24 TMUs, and only 8 ROPs. Those structural advantages compound under Vulkan's lower-level API overhead, where the hardware has to do more of the work directly.

The pixel and texture rates confirm the pattern. The NVIDIA card pushes 16.51 GPixel/s and 66.05 GTexel/s, while the AMD card manages 7.680 GPixel/s and 23.04 GTexel/s. Those are not close numbers. The GTX 650 Ti Boost delivers more than double the pixel throughput and nearly triple the texture fill rate. The FP32 compute figures follow suit: 1.585 TFLOPS for NVIDIA versus 737.3 GFLOPS for AMD, again a greater than 2:1 ratio.

Memory bandwidth also favors NVIDIA decisively. The GTX 650 Ti Boost runs a 192-bit bus with 144.2 GB/s of bandwidth, while the R7 M370 uses a 128-bit bus at 57.60 GB/s. That 2.5x bandwidth advantage matters in both compute and gaming workloads, especially at higher resolutions where memory traffic becomes the bottleneck.

The R7 M370's best case is its efficiency profile. It draws less power by virtue of its smaller die and lower clock speeds, though the database does not list a TDP for the AMD card. The GTX 650 Ti Boost is rated at 134 W. But in raw performance, the AMD card has no answer. Every recorded metric, from compute scores to fill rates to bandwidth, lands in NVIDIA's column.

Specification Differences

The two cards diverge on nearly every measurable specification. The GTX 650 Ti Boost uses a 192-bit memory bus, while the R7 M370 uses a 128-bit bus. Memory bandwidth reflects that difference: 144.2 GB/s versus 57.60 GB/s. Both cards carry 2 GB of GDDR5 memory, but the NVIDIA card's memory clock runs at 1502 MHz (6 Gbps effective) compared to 900 MHz (3.6 Gbps effective) for AMD.

Shading units: 768 versus 384. Texture mapping units: 64 versus 24. ROPs: 24 versus 8. The GTX 650 Ti Boost's pixel rate is 16.51 GPixel/s versus 7.680 GPixel/s. Its texture rate is 66.05 GTexel/s versus 23.04 GTexel/s. FP32 compute is 1.585 TFLOPS versus 737.3 GFLOPS.

The bus interface differs too. NVIDIA uses PCIe 3.0 x16, while AMD uses PCIe 3.0 x8. That halves the available bandwidth to the host system for the AMD card, though in practice most workloads do not saturate even x8.

Clock speeds are closer than the other specs. The GTX 650 Ti Boost runs at 980 MHz base and 1032 MHz boost. The R7 M370 runs at 875 MHz base and 960 MHz boost. NVIDIA's clocks are higher, but not dramatically so. The performance gap comes primarily from the larger chip and wider execution resources, not from clock speed alone.

The NVIDIA card has a 134 W TDP, 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 AMD card lists no TDP, no slot width, no power connector, no suggested PSU, and no dimensions. The NVIDIA card also specifies its display outputs: 2x DVI, 1x HDMI 1.4a, and 1x DisplayPort 1.2. The AMD card lists no display outputs.

DirectX support favors AMD slightly. The R7 M370 supports DirectX 12 (11_1), while the GTX 650 Ti Boost supports DirectX 12 (11_0). Both support OpenGL 4.6. Vulkan versions are close: 1.2.175 for NVIDIA and 1.2.170 for AMD. The NVIDIA card has a Geekbench Metal score of 4858; the AMD card has no Metal result.

Architecture Differences

The GTX 650 Ti Boost is built on NVIDIA's Kepler architecture, using the GK106 chip fabricated on TSMC's 28 nm process. The die measures 221 mm² and contains 2,540 million transistors, for a transistor density of 11.5M per mm². Kepler was designed for efficiency and scalability, and this chip shows that focus with its 768 shading units arranged across the GK106's execution blocks.

The R7 M370 uses AMD's GCN 1.0 architecture, the first iteration of Graphics Core Next. Its chip, codenamed Litho, is also on TSMC's 28 nm process but is far smaller: 77 mm² with 950 million transistors, yielding a slightly higher transistor density of 12.3M per mm². GCN 1.0 was AMD's shift toward a more compute-oriented design, but the Litho chip is a cut-down implementation with only 384 shading units.

The architectural philosophies differ. Kepler emphasizes raw throughput with many small cores and a robust memory subsystem. GCN 1.0 emphasizes wavefront-based execution and general-purpose compute, but the R7 M370's implementation is too resource-starved to realize that potential. The 2:1 gap in shading units, the 8 versus 24 ROP count, and the 24 versus 64 TMU count all stem from this fundamental difference in chip scale.

The transistor budgets tell the same story. NVIDIA packs 2,540 million transistors into its larger die, while AMD fits 950 million into a much smaller one. The GTX 650 Ti Boost belongs to the GeForce 600 generation, released in March 2013. The R7 M370 is part of AMD's Gem System (R7 M300) generation, released in May 2015. The AMD card is newer by over two years, yet it trails in every performance metric. The database shows that architectural efficiency gains did not compensate for the smaller chip and narrower memory path.

The NVIDIA card's memory clock runs at 1502 MHz with 6 Gbps effective data rate. The AMD card's memory runs at 900 MHz with 3.6 Gbps effective. Combined with the 192-bit versus 128-bit bus, the NVIDIA card achieves 144.2 GB/s against 57.60 GB/s. That 2.5x bandwidth advantage is a direct consequence of both wider bus and faster memory clock.

The Verdict

The data points to one conclusion: the NVIDIA GeForce GTX 650 Ti Boost is the substantially faster GPU. It wins both head-to-head benchmarks, holds a 31.7% lead in OpenCL, and a 55.3% lead in Vulkan. Its average benchmark score of 8067 places it in the 42nd percentile of all GPUs, four percentile points above the R7 M370's 38th.

For compute workloads, the choice is unambiguous. The GTX 650 Ti Boost delivers 1.585 TFLOPS of FP32 performance, more than double the R7 M370's 737.3 GFLOPS. Its 144.2 GB/s of memory bandwidth dwarfs the AMD card's 57.60 GB/s. Its 768 shading units and 64 TMUs provide the execution resources that the AMD card's 384 and 24 simply cannot match.

The R7 M370 is not without merit. It is a newer design, released in 2015 versus 2013, and it supports DirectX 12 (11_1) compared to the NVIDIA card's 12 (11_0). Its smaller die and lower power draw make it a plausible choice for compact or power-constrained systems, though the database does not list a TDP to quantify that advantage. But for anyone choosing between these two based on performance, the GTX 650 Ti Boost wins on every recorded metric.

Users who need the highest compute throughput, the widest memory bandwidth, or the best Vulkan performance should select the NVIDIA card. Users who prioritize a newer architecture generation or DirectX 12 feature level 11_1 support might consider the AMD card, but they should do so knowing that the performance gap is large and consistent. The verdict from the database is clear: the GTX 650 Ti Boost is the stronger GPU, and the R7 M370 trails by margins that no driver update or workload tuning is likely to close.