NVIDIA GeForce GTX 460 vs NVIDIA GeForce GTX 650 Ti Comparison

The Verdict

The data positions the NVIDIA GeForce GTX 650 Ti and the NVIDIA GeForce GTX 460 as extremely close performers, but the verdict splits cleanly by workload and platform priorities. The GTX 460 wins the only shared benchmark — Geekbench OpenCL — by a narrow 0.6% margin, posting a score of 7925 against the GTX 650 Ti’s 7877. That makes the older Fermi card the raw compute winner in this direct comparison, but the GTX 650 Ti counters with a broader feature set and significantly lower power demands.

For users prioritizing modern API support and efficiency, the GTX 650 Ti is the clear pick. It supports Vulkan 1.2.175, while the GTX 460 has no Vulkan support listed. The GTX 650 Ti also carries a 110 W TDP versus the GTX 460’s 160 W, requires a single 6-pin power connector instead of two, and asks for a 300 W suggested PSU rather than 450 W. It is also physically shorter at 145 mm (5.7 inches) versus 210 mm (8.3 inches) and occupies a single slot compared to the GTX 460’s dual-slot design.

For users who only care about the raw OpenCL number and already have the power delivery and chassis space, the GTX 460 edges ahead. Its 7925 score beats the GTX 650 Ti’s 7877, and its percentile ranking (41st) is only one point lower than the GTX 650 Ti’s 42nd percentile. The data shows the GTX 460 also has a 24 ROPS count versus 16, and a wider 192-bit memory bus versus 128-bit, though both cards deliver identical 86.40 GB/s bandwidth. The GTX 650 Ti is the modern choice; the GTX 460 is the narrow compute winner with a smaller feature footprint.

Where Each One Wins

The GTX 460 claims the only direct head-to-head victory in the dataset: Geekbench OpenCL, where it scores 7925 against the GTX 650 Ti’s 7877, a delta of -0.6% from the GTX 650 Ti’s perspective. That single win gives the GTX 460 a 1–0 record in the head-to-head benchmarks. The GTX 460’s advantage here likely stems from its higher ROPS count (24 versus 16) and wider memory bus (192-bit versus 128-bit), though both cards share the same 86.40 GB/s bandwidth.

The GTX 650 Ti wins in every category that does not involve raw compute throughput. It has a higher average benchmark score of 8053 across two tests (Geekbench OpenCL at 7877 and Geekbench Vulkan at 8229), whereas the GTX 460 only has one recorded benchmark. The Vulkan score of 8229 is particularly notable because the GTX 460 cannot run that test at all — the data lists no Vulkan API support for the Fermi card. The GTX 650 Ti also wins decisively on efficiency: 110 W TDP versus 160 W, single 6-pin connector versus dual 6-pin, 300 W suggested PSU versus 450 W, and single-slot versus dual-slot footprint. It is also shorter at 145 mm versus 210 mm, making it the better fit for compact systems. The GTX 650 Ti’s transistor density is nearly double at 11.5M / mm² versus 5.9M / mm², achieved on a 28 nm process versus 40 nm.

Architecture Differences

The two cards come from entirely different NVIDIA architectures. The GTX 650 Ti is built on Kepler (chip GK106S), while the GTX 460 uses Fermi (chip GF104). Kepler is the newer generation (GeForce 600) compared to Fermi (GeForce 400), and the process node reflects that: 28 nm versus 40 nm, both fabricated by TSMC. The GTX 650 Ti packs 2,540 million transistors into a 221 mm² die, while the GTX 460 has 1,950 million transistors on a much larger 332 mm² die. This gives the GTX 650 Ti a transistor density of 11.5M / mm² versus 5.9M / mm² for the GTX 460.

Shader configuration differs substantially. The GTX 650 Ti has 768 shading units and 64 TMUs, while the GTX 460 has 336 shading units and 56 TMUs. The GTX 460 counters with more ROPs: 24 versus 16. Pixel rate favors the GTX 650 Ti at 14.85 GPixel/s versus 9.450 GPixel/s, and texture rate is also higher at 59.39 GTexel/s versus 37.80 GTexel/s. FP32 performance is 1,425.4 GFLOPS for the GTX 650 Ti versus 907.2 GFLOPS for the GTX 460 — a 57% advantage for the Kepler card in raw floating-point throughput. Neither card has RT cores or tensor cores, and neither lists FP16 performance.

Memory subsystems are similar in bandwidth but different in configuration. Both cards use GDDR5 and deliver exactly 86.40 GB/s, but the GTX 650 Ti achieves this with 1024 MB on a 128-bit bus at 1350 MHz (5.4 Gbps effective), while the GTX 460 uses 768 MB on a 192-bit bus at 900 MHz (3.6 Gbps effective). API support diverges: both support DirectX 12 (11_0) and OpenGL 4.6, but the GTX 650 Ti adds Vulkan 1.2.175 while the GTX 460 lists no Vulkan support. Power delivery and physical design also differ: the GTX 650 Ti runs on 110 W with a single 6-pin connector and single-slot cooler, while the GTX 460 draws 160 W with two 6-pin connectors and a dual-slot cooler. The GTX 650 Ti uses PCIe 3.0 x16, while the GTX 460 uses PCIe 2.0 x16.

FAQ

Q: Which card has the higher average benchmark score?

A: The NVIDIA GeForce GTX 650 Ti has an average benchmark score of 8053, based on two Geekbench tests (OpenCL at 7877 and Vulkan at 8229). The NVIDIA GeForce GTX 460 has an average score of 7925, based only on its Geekbench OpenCL result.

Q: Does the GTX 460 beat the GTX 650 Ti in any benchmark?

A: Yes. In Geekbench OpenCL, the GTX 460 scores 7925 against the GTX 650 Ti’s 7877, a 0.6% margin in favor of the GTX 460. This is the only shared benchmark, and it gives the GTX 460 a 1–0 head-to-head win record.

Q: Why does the GTX 650 Ti have a higher FP32 rating despite losing in OpenCL?

A: The GTX 650 Ti is rated at 1,425.4 GFLOPS FP32 versus 907.2 GFLOPS for the GTX 460, a 57% theoretical advantage. However, the actual Geekbench OpenCL result shows the GTX 460 winning by 0.6%, indicating that theoretical throughput does not translate directly to this particular compute workload. The GTX 460’s higher ROP count (24 versus 16) and wider memory bus (192-bit versus 128-bit) may offset some of the shader advantage.

Q: Can the GTX 460 run Vulkan applications?

A: The data lists no Vulkan support for the GTX 460. The GTX 650 Ti supports Vulkan 1.2.175. Both cards support DirectX 12 (11_0) and OpenGL 4.6.

Q: Which card requires less power and cooling?

A: The GTX 650 Ti is far more efficient. It has a 110 W TDP, needs a single 6-pin power connector, and a 300 W suggested PSU. The GTX 460 has a 160 W TDP, requires two 6-pin connectors, and a 450 W suggested PSU. The GTX 650 Ti is also single-slot and 145 mm long, while the GTX 460 is dual-slot and 210 mm long.

Q: How do the two cards compare in memory capacity?

A: The GTX 650 Ti has 1024 MB of GDDR5 on a 128-bit bus, while the GTX 460 has 768 MB on a 192-bit bus. Both deliver identical bandwidth at 86.40 GB/s.

Head-to-Head Benchmarks

The only head-to-head benchmark in the dataset is Geekbench OpenCL. The GTX 460 wins with a score of 7925 against the GTX 650 Ti’s 7877, producing a delta of -0.6% from the GTX 650 Ti’s perspective. This is a razor-thin margin — effectively a tie in real-world terms, but it does give the GTX 460 the only direct win in the comparison. The GTX 650 Ti’s loss here is surprising given its theoretical advantages: it has more than double the shading units (768 versus 336), a 57% higher FP32 rating (1,425.4 versus 907.2 GFLOPS), and a higher texture rate (59.39 versus 37.80 GTexel/s). The GTX 460’s win likely comes down to its higher ROP count (24 versus 16) and wider memory bus (192-bit versus 128-bit), which may favor certain compute patterns despite identical bandwidth.

Beyond that single shared test, the GTX 650 Ti demonstrates its strength in the Vulkan benchmark, scoring 8229 — a result the GTX 460 cannot produce because it lacks Vulkan support entirely. This 8229 score is the highest individual benchmark result for either card, and it pushes the GTX 650 Ti’s average to 8053 versus 7925 for the GTX 460. The GTX 650 Ti’s percentile ranking of 42 versus the GTX 460’s 41 reflects this broader capability. In the nearest-rival context, the GTX 650 Ti’s 8053 average sits just 0.2% above the GTX 880M (8040) and Quadro P5000 (8039), and 0.2% below the GTX 650 Ti Boost (8067). The GTX 460’s 7925 average is 0.2% above the Quadro K4100M (7906) and 1.4% below both the Quadro P5000 and GTX 880M, with the GTX 650 Ti sitting 1.6% above it.

The biggest win for the GTX 650 Ti is not a single benchmark but the aggregate capability: it offers Vulkan performance where the GTX 460 has none, and it does so while drawing 50 W less power (110 W versus 160 W) and using half the power connectors. The biggest win for the GTX 460 is its OpenCL score, which edges out the newer card despite being released earlier (July 2010 versus October 2012) and using an older architecture.

Specification Differences

The two cards differ across nearly every major specification category. The GTX 650 Ti uses a 28 nm process with 2,540 million transistors on a 221 mm² die (11.5M / mm² density), while the GTX 460 uses a 40 nm process with 1,950 million transistors on a 332 mm² die (5.9M / mm² density). Memory clocks differ: the GTX 650 Ti runs at 1350 MHz (5.4 Gbps effective), while the GTX 460 runs at 900 MHz (3.6 Gbps effective). Memory capacity is 1024 MB for the GTX 650 Ti versus 768 MB for the GTX 460, with bus widths of 128-bit and 192-bit respectively — though bandwidth is identical at 86.40 GB/s.

Shader resources differ sharply: 768 shading units and 64 TMUs for the GTX 650 Ti versus 336 shading units and 56 TMUs for the GTX 460. ROPs go the other way: 16 for the GTX 650 Ti versus 24 for the GTX 460. Pixel rate is 14.85 GPixel/s versus 9.450 GPixel/s, and texture rate is 59.39 GTexel/s versus 37.80 GTexel/s. FP32 is 1,425.4 GFLOPS versus 907.2 GFLOPS. TDP is 110 W versus 160 W, with power connectors of 1x 6-pin versus 2x 6-pin, and suggested PSU of 300 W versus 450 W. Slot width is single-slot versus dual-slot, and length is 145 mm versus 210 mm. Bus interface is PCIe 3.0 x16 versus PCIe 2.0 x16. Display outputs are 2x DVI and 1x mini-HDMI 1.4a versus 2x DVI and 1x mini-HDMI 1.3a. API support differs on Vulkan: 1.2.175 for the GTX 650 Ti, none for the GTX 460; both share DirectX 12 (11_0) and OpenGL 4.6. The GTX 650 Ti’s launch MSRP was 149 USD, while the GTX 460’s was 199 USD.