NVIDIA GeForce GTX 460 vs NVIDIA GeForce GTX 660 Comparison

Head-to-Head Benchmarks

The only directly comparable benchmark in the database is Geekbench OpenCL, and the result is decisive. The NVIDIA GeForce GTX 660 scores 11,347 points, while the NVIDIA GeForce GTX 460 scores 7,925 points. That is a delta of 43.2% in favor of the GTX 660. This is not a marginal generational bump; it is a substantial leap in raw compute throughput, and the architecture shift between the two largely explains it.

Looking at the broader database context, the GTX 660 sits at the 45th percentile among all GPUs, with an average benchmark score of 9,022. Its nearest rivals include the NVIDIA TITAN V CEO Edition at 9,037 (only 0.2% ahead), the NVIDIA GeForce GTX 560 at 9,058 (0.4% ahead), the AMD Radeon 550X at 8,918 (1.2% behind), and the AMD Radeon Pro WX 5100 at 8,863 (1.8% behind). The GTX 660 is effectively clustered with much newer and more expensive hardware, which is remarkable for a card from the GeForce 600 generation.

The GTX 460, by contrast, sits at the 41st percentile with an average score of 7,925. Its nearest rivals are the NVIDIA Quadro K4100M at 7,906 (0.2% behind the GTX 460), the NVIDIA Quadro P5000 at 8,039 (1.4% ahead), the NVIDIA GeForce GTX 880M at 8,040 (1.4% ahead), and the NVIDIA GeForce GTX 650 Ti at 8,053 (1.6% ahead). The GTX 460 is competitive with mobile and entry-level workstation parts from later generations, but it cannot touch the GTX 660.

The 43.2% OpenCL delta is the single biggest head-to-head win in the dataset. There is no benchmark where the GTX 460 comes out ahead. The GTX 660 wins the only shared test, and it does so by a wide margin that reflects both a higher shader count and a more efficient architecture.

The Verdict

The data is unambiguous: the NVIDIA GeForce GTX 660 outperforms the NVIDIA GeForce GTX 460 in the only directly comparable benchmark. The GTX 660 scores 11,347 versus 7,925 in Geekbench OpenCL, a 43.2% advantage. The GTX 660 also ranks higher overall, sitting at the 45th percentile versus the GTX 460's 41st percentile.

For anyone choosing between these two, the GTX 660 is the stronger pick across every measurable dimension in this database. It delivers more compute performance, higher memory bandwidth, and better API support. The GTX 460 is an older design from the GeForce 400 generation, and while it was a capable card in its time, the recorded data shows it has been surpassed by a significant margin.

The GTX 660's average benchmark score of 9,022 places it in the company of cards like the NVIDIA TITAN V CEO Edition and the AMD Radeon Pro WX 5100, which confirms how well it has aged in raw compute terms. The GTX 460's average of 7,925 places it closer to the NVIDIA Quadro K4100M and the GTX 650 Ti, both of which are newer or more specialized parts.

There is no scenario in the data where the GTX 460 wins. The GTX 660 wins the head-to-head, holds a higher percentile, and offers a more modern feature set. The verdict is straightforward.

Where Each One Wins

The GTX 660 wins in every category where the database provides a direct comparison. In Geekbench OpenCL, it is 43.2% ahead. In terms of memory bandwidth, the GTX 660 offers 144.2 GB/s versus 86.40 GB/s for the GTX 460, a difference of roughly 67%. The GTX 660 also has a faster effective memory clock at 6 Gbps versus 3.6 Gbps.

The GTX 660's shading unit count is 960 versus 336 for the GTX 460, and its texture units number 80 versus 56. The pixel rate is 20.64 GPixel/s versus 9.450 GPixel/s, and the texture rate is 82.56 GTexel/s versus 37.80 GTexel/s. The FP32 compute throughput is 1.981 TFLOPS versus 907.2 GFLOPS, more than double.

The GTX 460 does have a few areas where it is not strictly worse, though the data does not translate those into wins. It has the same 192-bit memory bus width as the GTX 660, and the same 24 ROPs. Its TDP is higher at 160 W versus 140 W for the GTX 660, which means the GTX 660 delivers more performance while drawing less power. The GTX 460 also requires a 450 W suggested PSU versus 300 W for the GTX 660.

In terms of API support, the GTX 660 adds Vulkan 1.2.175, while the GTX 460 lists no Vulkan support at all. Both support DirectX 12 (11_0) and OpenGL 4.6. The GTX 660 also has a more modern display output set, including DisplayPort 1.2, while the GTX 460 is limited to mini-HDMI 1.3a.

FAQ

Q: Which GPU is faster in OpenCL benchmarks?

A: The NVIDIA GeForce GTX 660 scores 11,347 in Geekbench OpenCL, while the NVIDIA GeForce GTX 460 scores 7,925. The GTX 660 is 43.2% faster in this test.

Q: How do the two cards compare in overall percentile ranking?

A: The GTX 660 ranks at the 45th percentile among all GPUs, while the GTX 460 ranks at the 41st percentile. The GTX 660 sits higher in the overall distribution.

Q: Do both cards support the same APIs?

A: No. Both support DirectX 12 (11_0) and OpenGL 4.6, but the GTX 660 also supports Vulkan 1.2.175, while the GTX 460 lists no Vulkan support.

Q: What is the memory bandwidth difference?

A: The GTX 660 has 144.2 GB/s of memory bandwidth, while the GTX 460 has 86.40 GB/s. The GTX 660 offers roughly 67% more bandwidth.

Q: Which card has higher power requirements?

A: The GTX 460 has a TDP of 160 W and a suggested PSU of 450 W, while the GTX 660 has a TDP of 140 W and a suggested PSU of 300 W. The GTX 460 requires more power despite lower performance.

Q: Are there any benchmarks where the GTX 460 wins?

A: No. In the only head-to-head benchmark recorded, Geekbench OpenCL, the GTX 660 wins by 43.2%. The GTX 460 has no wins in the database.

Architecture Differences

The two GPUs come from different architectural generations. The GTX 660 is built on the Kepler architecture with the GK106 chip, while the GTX 460 uses the Fermi architecture with the GF104 chip. This is a fundamental generational shift, and it shows up in nearly every specification.

The manufacturing process differs substantially. The GTX 660 is fabricated on a 28 nm process at TSMC, while the GTX 460 uses a 40 nm process, also at TSMC. The smaller node allows the GTX 660 to pack 2,540 million transistors into a die size of 221 mm², giving it a transistor density of 11.5M per mm². The GTX 460 has 1,950 million transistors on a much larger die of 332 mm², resulting in a density of only 5.9M per mm². The GTX 660 achieves higher transistor density and a smaller physical die, which contributes to its lower TDP of 140 W versus 160 W.

The shader architecture is also very different. The GTX 660 has 960 shading units, 80 TMUs, and 24 ROPs. The GTX 460 has 336 shading units, 56 TMUs, and 24 ROPs. The ROP count is identical, which means the GTX 660's higher pixel rate comes entirely from its higher clock speeds. The GTX 660 runs at a base clock of 980 MHz with a boost of 1032 MHz, while the GTX 460 has no listed base or boost clock in the database. The GTX 660's memory runs at 1502 MHz (6 Gbps effective), while the GTX 460's memory runs at 900 MHz (3.6 Gbps effective).

The bus interface also differs. The GTX 660 uses PCIe 3.0 x16, while the GTX 460 uses PCIe 2.0 x16. Both cards are dual-slot designs, but the GTX 660 uses a single 6-pin power connector while the GTX 460 requires two 6-pin connectors. The GTX 660 is also longer at 241 mm (9.5 inches) versus 210 mm (8.3 inches) for the GTX 460.

Specification Differences

The memory configuration differs in size but not bus width. The GTX 660 has 2 GB of GDDR5 on a 192-bit bus, while the GTX 460 has 768 MB of GDDR5 on the same 192-bit bus. The bandwidth difference is significant: 144.2 GB/s for the GTX 660 versus 86.40 GB/s for the GTX 460.

The compute specifications are starkly different. The GTX 660 delivers 1.981 TFLOPS of FP32 performance, a pixel rate of 20.64 GPixel/s, and a texture rate of 82.56 GTexel/s. The GTX 460 delivers 907.2 GFLOPS, 9.450 GPixel/s, and 37.80 GTexel/s. The GTX 660 is more than twice as fast in raw FP32 throughput.

The power profile favors the newer card. The GTX 660 has a TDP of 140 W and a suggested PSU of 300 W, while the GTX 460 has a TDP of 160 W and a suggested PSU of 450 W. The GTX 660 also requires fewer power connectors: one 6-pin versus two 6-pin.

Display outputs differ as well. The GTX 660 offers 2x DVI, 1x HDMI 1.4a, and 1x DisplayPort 1.2. The GTX 460 offers 2x DVI and 1x mini-HDMI 1.3a. The GTX 660 adds DisplayPort support and a newer HDMI revision.

The release dates are separated by more than two years: the GTX 460 launched in July 2010, while the GTX 660 launched in September 2012. The GTX 460's predecessor is the GeForce 200 series and its successor is the GeForce 500 series. The GTX 660's predecessor is the GeForce 500 series and its successor is the GeForce 700 series. Both cards are marked as end-of-life in the database. The GTX 660 launched with an MSRP of 229 USD, while the GTX 460 launched with an MSRP of 199 USD.