AMD Radeon R9 M265X vs NVIDIA GeForce GTX 650 Ti Comparison

The Verdict

The recorded data separates these two end-of-life graphics cards by a single benchmark result, the Geekbench OpenCL test, and the margin is decisive. The AMD Radeon R9 M265X wins that head-to-head with a score of 8851 against the NVIDIA GeForce GTX 650 Ti's 7877, a 12.4% advantage. The GTX 650 Ti has a second recorded benchmark, a Geekbench Vulkan score of 8229, but no comparable Vulkan result exists for the R9 M265X in the database, so that test cannot be used for a direct comparison.

Looking at the broader percentile rankings, the R9 M265X sits at the 44th percentile of all GPUs, while the GTX 650 Ti sits at the 42nd. The average benchmark score tells the same story: the AMD part averages 8851, the NVIDIA part averages 8053. The nearest rivals for each card confirm their relative positions. The R9 M265X trades blows with the AMD Radeon Pro WX 5100 (8863, a 0.1% gap), the AMD Radeon 550X (8918, a 0.8% gap), the NVIDIA GeForce RTX 3050 A Mobile (8746, a 1.2% gap), and the NVIDIA GeForce GTX 460 v2 (8743, a 1.2% gap). The GTX 650 Ti, meanwhile, is clustered with the NVIDIA GeForce GTX 880M (8040, a 0.2% gap), the GTX 650 Ti Boost (8067, a 0.2% gap), the Quadro P5000 (8039, a 0.2% gap), and the GRID K2 (8080, a 0.3% gap).

The verdict is straightforward: the R9 M265X is the faster card in the only directly comparable workload. It also carries twice the memory, 2 GB versus 1024 MB, which matters for texture-heavy scenes. The GTX 650 Ti, however, has a much higher raw compute ceiling in its specifications, which the benchmark does not reflect. For a buyer choosing strictly from this data, the R9 M265X is the pick for OpenCL performance and memory capacity. The GTX 650 Ti is the pick only if Vulkan support is a requirement, because it has a recorded Vulkan score while the AMD part has none.

Where Each One Wins

The head-to-head table shows a single victory for the AMD Radeon R9 M265X in the Geekbench OpenCL test. That is the only win recorded for either card, so the AMD part takes the sole direct benchmark crown. The 12.4% delta is not trivial; it places the R9 M265X clearly above the GTX 650 Ti in compute workloads that use OpenCL.

The GTX 650 Ti has no direct wins, but it does have a Vulkan score of 8229 in the database. Since the R9 M265X has no Vulkan result recorded, the NVIDIA card holds the advantage in that API by default. The GTX 650 Ti also has the higher specification ceiling: 768 shading units against 640, 64 texture mapping units against 40, a texture rate of 59.39 GTexel/s against 25.00 GTexel/s, and an FP32 throughput of 1,425.4 GFLOPS against 800.0 GFLOPS. Those numbers suggest the NVIDIA card should win in fill-rate-bound or shader-heavy scenarios, but the database does not contain a benchmark that proves it.

In terms of memory bandwidth, the GTX 650 Ti wins with 86.40 GB/s versus 64.00 GB/s, a 35% advantage. The AMD card wins on memory capacity with 2 GB versus 1024 MB. For larger textures or higher resolutions, the capacity advantage is practical; for raw data throughput, the bandwidth advantage is practical. The pixel rate also favors NVIDIA, 14.85 GPixel/s versus 10.00 GPixel/s, which matters for rasterization at high resolutions.

Architecture Differences

The two GPUs come from different architectural generations and different manufacturers. The AMD Radeon R9 M265X uses the Venus chip built on GCN 1.0, while the NVIDIA GeForce GTX 650 Ti uses the GK106S chip built on Kepler. Both are fabricated on a 28 nm process at TSMC, but the transistor counts differ sharply. The NVIDIA chip packs 2,540 million transistors on a 221 mm² die, while the AMD chip uses 1,500 million transistors on a 123 mm² die. The transistor density is similar, 11.5M per mm² for NVIDIA and 12.2M per mm² for AMD, but the NVIDIA die is nearly twice the physical size.

The release dates are separated by roughly 18 months. The GTX 650 Ti launched on October 8, 2012, while the R9 M265X launched on March 20, 2014. The AMD part belongs to the Gem System generation within the R9 M200 family, and its predecessor is listed as Solar System with a successor of Polaris Mobile. The NVIDIA part belongs to the GeForce 600 generation, with a predecessor of GeForce 500 and a successor of GeForce 700.

Core configuration differences are significant. The GTX 650 Ti has 768 shading units, 64 TMUs, and 16 ROPs. The R9 M265X has 640 shading units, 40 TMUs, and 16 ROPs. The NVIDIA part therefore has 20% more shaders, 60% more TMUs, and the same ROP count. The memory buses are both 128 bit, but the GTX 650 Ti runs its GDDR5 at a higher effective speed, 5.4 Gbps versus 4 Gbps, producing the bandwidth gap noted earlier.

API support differs in one key way. The AMD part supports DirectX 12 (11_1), while the NVIDIA part supports DirectX 12 (11_0). Both support OpenGL 4.6. For Vulkan, the NVIDIA part records version 1.2.175, while the AMD part records 1.2.170. The NVIDIA card also has a recorded Vulkan benchmark, the AMD card does not.

FAQ

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

A: The AMD Radeon R9 M265X scores 8851, which is 12.4% higher than the NVIDIA GeForce GTX 650 Ti's 7877.

Q: Does the NVIDIA card have any benchmark win?

A: No. In the head-to-head comparison, the AMD card wins the only directly comparable test, Geekbench OpenCL. The NVIDIA card has a separate Geekbench Vulkan score of 8229, but the AMD card has no Vulkan result recorded.

Q: How much memory does each card have?

A: The AMD Radeon R9 M265X has 2 GB of GDDR5, while the NVIDIA GeForce GTX 650 Ti has 1024 MB of GDDR5. Both use a 128 bit memory bus.

Q: Which card has higher memory bandwidth?

A: The NVIDIA GeForce GTX 650 Ti, at 86.40 GB/s, compared to 64.00 GB/s for the AMD Radeon R9 M265X.

Q: What is the transistor count difference?

A: The NVIDIA GK106S chip contains 2,540 million transistors on a 221 mm² die. The AMD Venus chip contains 1,500 million transistors on a 123 mm² die.

Q: Which card supports a higher DirectX version?

A: The AMD Radeon R9 M265X supports DirectX 12 (11_1), while the NVIDIA GeForce GTX 650 Ti supports DirectX 12 (11_0).

Head-to-Head Benchmarks

The database records exactly one head-to-head benchmark between these two cards: Geekbench OpenCL. The AMD Radeon R9 M265X scores 8851, the NVIDIA GeForce GTX 650 Ti scores 7877. The delta is 12.4% in favor of AMD. That is a substantial margin for two cards that sit within two percentile points of each other in the global ranking (44th versus 42nd). The average benchmark scores amplify the gap: the AMD part averages 8851, the NVIDIA part averages 8053, a difference of 798 points.

The NVIDIA card's Vulkan score of 8229 is worth examining. It falls between the two OpenCL scores, closer to the GTX 650 Ti's OpenCL result than to the AMD's OpenCL result. If the R9 M265X had a Vulkan result, the comparison would be more complete, but the database lacks that data point. The nearest rivals for the GTX 650 Ti show a tight cluster: the GTX 880M at 8040, the GTX 650 Ti Boost at 8067, the Quadro P5000 at 8039, and the GRID K2 at 8080. All are within 0.3% of the GTX 650 Ti's average. The R9 M265X's rivals similarly cluster: the Pro WX 5100 at 8863, the Radeon 550X at 8918, the RTX 3050 A Mobile at 8746, and the GTX 460 v2 at 8743. The AMD part's nearest rivals are all within 1.2%, but that cluster sits roughly 10% higher than the NVIDIA card's cluster.

The specification sheet predicts a different outcome than the benchmark. The GTX 650 Ti has higher FP32 throughput (1,425.4 GFLOPS versus 800.0 GFLOPS), higher texture rate (59.39 GTexel/s versus 25.00 GTexel/s), and higher pixel rate (14.85 GPixel/s versus 10.00 GPixel/s). Yet the OpenCL test gives the win to AMD. This suggests the workload favors the AMD architecture's compute scheduling or memory capacity rather than raw shader count. The 2 GB memory buffer on the AMD card likely helps in OpenCL workloads that allocate large buffers.

Specification Differences

The two cards differ across nearly every specification field. The process node is identical, 28 nm at TSMC, and both use a PCIe 3.0 x16 bus interface. The memory type is identical, GDDR5, and the bus width is identical, 128 bit. Beyond that, the differences are extensive.

The AMD Radeon R9 M265X has 640 shading units, 40 TMUs, and 16 ROPs. The NVIDIA GeForce GTX 650 Ti has 768 shading units, 64 TMUs, and 16 ROPs. The NVIDIA part offers more shading and texture hardware but the same rasterization hardware.

Clock speeds differ in format. The AMD card has a base clock of 575 MHz and a boost clock of 625 MHz. The NVIDIA card lists no base or boost clock, only a memory clock of 1350 MHz with an effective data rate of 5.4 Gbps. The AMD memory clock is 1000 MHz with a 4 Gbps effective rate. The resulting bandwidth is 86.40 GB/s for NVIDIA and 64.00 GB/s for AMD.

Memory capacity favors AMD, 2 GB versus 1024 MB. The GTX 650 Ti has a TDP of 110 W, a single-slot design, a 6-pin power connector, and a suggested PSU of 300 W. The AMD card lists no TDP, no slot width, no power connectors, and no suggested PSU. The GTX 650 Ti measures 145 mm or 5.7 inches in length. The AMD card has no recorded dimensions.

The GTX 650 Ti has display outputs of 2x DVI and 1x mini-HDMI 1.4a. The AMD card lists no display outputs. The NVIDIA card has a launch MSRP of 149 USD. The AMD card has no recorded MSRP.

API support differs in DirectX version and Vulkan version. The AMD card supports DirectX 12 (11_1) and Vulkan 1.2.170. The NVIDIA card supports DirectX 12 (11_0) and Vulkan 1.2.175. Both support OpenGL 4.6.

The die sizes and transistor counts are the largest physical difference. The NVIDIA GK106S die is 221 mm² with 2,540 million transistors, a density of 11.5M per mm². The AMD Venus die is 123 mm² with 1,500 million transistors, a density of 12.2M per mm². The AMD die is 44% smaller but achieves a slightly higher transistor density. The NVIDIA card was released on October 8, 2012, the AMD card on March 20, 2014. Both are end-of-life products.