NVIDIA GeForce GTX 460 vs NVIDIA GeForce GTX 970 Comparison

Head-to-Head Benchmarks

The recorded data contains a single shared benchmark between these two GPUs: Geekbench OpenCL. In that test, the NVIDIA GeForce GTX 970 delivers a score of 29,982, while the NVIDIA GeForce GTX 460 scores 7,925. The delta between them is 73.6% in favor of the GTX 970, meaning the newer card is roughly 3.8 times faster in raw compute throughput as measured by this workload. This is not a marginal generational step; it is a decisive gap that places the two cards in entirely different performance tiers.

The GTX 970’s OpenCL result is strong enough to sit well above its nearest rivals in the database. Its nearest competitor, the Intel Iris Pro Graphics P580, averages 7,170, which is 0.2% below the GTX 970’s average score of 7,157 across all its recorded benchmarks. Note that the OpenCL score of 29,982 is far higher than the GTX 970’s overall average, which is dragged down by other tests in its suite. The GTX 460, by contrast, has a single recorded benchmark, so its average score of 7,925 is identical to its OpenCL result. The GTX 460’s nearest rivals are tightly clustered: the NVIDIA Quadro K4100M at 7,906 (0.2% below), the NVIDIA Quadro P5000 at 8,039 (1.4% above), the NVIDIA GeForce GTX 880M at 8,040 (1.4% above), and the NVIDIA GeForce GTX 650 Ti at 8,053 (1.6% above). The GTX 460 is therefore competitive with that group, but it is nowhere near the GTX 970’s compute output.

The single head-to-head result shows a clear winner. The GTX 970 wins the only shared test, giving it a 1-0 record in direct comparison. The GTX 460 has no wins in any recorded head-to-head matchup. The delta of 73.6% is large enough that no amount of driver tuning or overclocking could realistically close the gap in this workload. The OpenCL test stresses raw shader throughput and memory bandwidth, both of which are heavily in the GTX 970’s favor based on the specification data.

Where Each One Wins

The GTX 970 wins the only benchmark the two cards share, so in terms of measured performance, it is the clear victor. The OpenCL result indicates strong general-purpose compute capability, which aligns with its much larger shading unit count and higher memory bandwidth. For workloads like GPU-accelerated rendering, physics simulation, or data-parallel processing, the GTX 970 would be the obvious choice based on this data.

The GTX 460, while losing the direct comparison, still holds its own among its own nearest rivals. Its score of 7,925 is within 1.6% of the GeForce GTX 650 Ti, which is a later-generation card. This suggests that for older DirectX 11-era games or legacy compute tasks, the GTX 460 remains a serviceable performer relative to its contemporaries. It also has a lower transistor count and smaller die size, which historically correlates with lower power draw per unit of performance, though no direct efficiency metric is recorded in the database.

For gaming specifically, the GTX 970 has a much broader benchmark suite recorded. Its Passmark scores show strengths across different DirectX versions: DirectX 9 at 143, DirectX 10 at 46, DirectX 11 at 71, and DirectX 12 at 41. The DirectX 9 score is notably higher than the others, which suggests strong legacy API performance. The DirectX 11 score of 71 is also respectable. The GTX 460 has no recorded gaming benchmarks, so its gaming performance must be inferred from its OpenCL result and its specification profile. The GTX 970 also records a Passmark G3D score of 9,638 and a compute score of 4,073, both of which are far beyond what the GTX 460 could likely achieve given its 907.2 GFLOPS FP32 throughput versus the GTX 970’s 3.920 TFLOPS.

In practical terms, the GTX 970 is the card for modern workloads: higher resolution textures, newer API features, and compute-heavy tasks. The GTX 460 is a legacy part, suitable for older software or as a budget secondary card, but the data shows no scenario where it outperforms the GTX 970.

FAQ

Q: What is the performance difference between the GTX 460 and GTX 970 in OpenCL?

A: The GTX 970 scores 29,982, while the GTX 460 scores 7,925. The GTX 970 is 73.6% faster, meaning it delivers roughly 3.8 times the performance in this test.

Q: Which GPU has more shading units?

A: The GTX 970 has 1,664 shading units, while the GTX 460 has 336. That is nearly a five-fold difference in shader count.

Q: What is the memory bandwidth of each card?

A: The GTX 970 has 224.4 GB/s of bandwidth, while the GTX 460 has 86.40 GB/s. The GTX 970 also has a wider 256-bit bus versus the GTX 460’s 192-bit bus, and 4 GB of memory versus 768 MB.

Q: How do the two cards compare in terms of pixel and texture rates?

A: The GTX 970 achieves 65.97 GPixel/s and 122.5 GTexel/s. The GTX 460 achieves 9.450 GPixel/s and 37.80 GTexel/s. The GTX 970 is substantially faster in both metrics.

Q: Which card supports newer DirectX versions?

A: The GTX 970 supports DirectX 12 (12_1), while the GTX 460 supports DirectX 12 (11_0). The GTX 970 also supports Vulkan 1.4, whereas the GTX 460 has no recorded Vulkan support.

Q: What is the average benchmark score for each card?

A: The GTX 460 has an average score of 7,925, and the GTX 970 has an average score of 7,157. However, the GTX 970’s average is pulled down by a wider range of tests; its OpenCL score is 29,982.

Specification Differences

The two cards differ across nearly every major specification. The GTX 460 uses a 40 nm process node, while the GTX 970 uses 28 nm. Transistor count is 1,950 million for the GTX 460 versus 5,200 million for the GTX 970. Die size is 332 mm² versus 398 mm², giving transistor densities of 5.9M per mm² and 13.1M per mm², respectively.

Memory configuration is another major split. The GTX 460 has 768 MB of GDDR5 on a 192-bit bus, with memory clocked at 900 MHz (3.6 Gbps effective) and bandwidth of 86.40 GB/s. The GTX 970 has 4 GB of GDDR5 on a 256-bit bus, with memory at 1753 MHz (7 Gbps effective) and bandwidth of 224.4 GB/s. The GTX 970 also has a base clock of 1050 MHz and a boost clock of 1178 MHz, while the GTX 460 has no recorded base or boost clocks.

Compute resources differ heavily. The GTX 460 has 336 shading units, 56 TMUs, and 24 ROPs. The GTX 970 has 1,664 shading units, 104 TMUs, and 56 ROPs. FP32 throughput is 907.2 GFLOPS for the GTX 460 versus 3.920 TFLOPS for the GTX 970. Pixel rate is 9.450 GPixel/s versus 65.97 GPixel/s, and texture rate is 37.80 GTexel/s versus 122.5 GTexel/s.

Power and physical specs also differ. The GTX 460 has a TDP of 160 W and a suggested PSU of 450 W. The GTX 970 has a TDP of 148 W and a suggested PSU of 300 W. Both are dual-slot cards with 2x 6-pin power connectors. The GTX 460 measures 210 mm (8.3 inches) in length, while the GTX 970 measures 267 mm (10.5 inches), with a height of 111 mm (4.4 inches) and width of 40 mm (1.6 inches). The GTX 460 uses PCIe 2.0 x16, while the GTX 970 uses PCIe 3.0 x16.

Display outputs differ as well. The GTX 460 has 2x DVI and 1x mini-HDMI 1.3a. The GTX 970 has 1x DVI, 1x HDMI 2.0, and 3x DisplayPort 1.2. API support also diverges: the GTX 460 lists DirectX 12 (11_0) and OpenGL 4.6, with no Vulkan. The GTX 970 lists DirectX 12 (12_1), OpenGL 4.6, and Vulkan 1.4.

Architecture Differences

The GTX 460 is built on the Fermi architecture, using the GF104 chip. The GTX 970 is built on the Maxwell 2.0 architecture, using the GM204 chip. These are separated by two full generations in NVIDIA’s lineup: the GTX 460 belongs to the GeForce 400 series, while the GTX 970 belongs to the GeForce 900 series. The GTX 460’s predecessor is the GeForce 200 series and its successor is the GeForce 500 series. The GTX 970’s predecessor is the GeForce 700 series and its successor is the GeForce 10 series.

The process node difference is significant: 40 nm for Fermi versus 28 nm for Maxwell 2.0. Both chips are fabricated by TSMC, but the newer 28 nm process allows for a much higher transistor density, 13.1M per mm² versus 5.9M per mm². This is why the GTX 970 can pack 5,200 million transistors on a die that is only slightly larger than the GTX 460’s 332 mm².

Architecturally, Maxwell 2.0 introduced major improvements in shading efficiency and power management. The GTX 970 achieves higher performance per watt despite having a lower TDP (148 W versus 160 W). The GTX 460’s Fermi architecture was known for its compute capability but was less efficient per clock. The GTX 970’s boost clock mechanism, with a base of 1050 MHz and boost of 1178 MHz, represents a later evolution of dynamic clocking that the GTX 460 does not have.

Memory architecture also reflects the generational gap. The GTX 970’s 256-bit bus and 4 GB capacity are designed for high-resolution gaming and large datasets, while the GTX 460’s 192-bit bus and 768 MB capacity were suited to the 2010-era game environment. The GTX 970’s support for DirectX 12 (12_1) includes features like conservative rasterization and raster order views, which are absent from the GTX 460’s DirectX 12 (11_0) profile. Vulkan support on the GTX 970 further extends its modern API compatibility.

Both cards are end-of-life in production status. The GTX 460 was released on July 11, 2010, and the GTX 970 on September 18, 2014. Their launch MSRPs were 199 USD and 329 USD, respectively. The GTX 970’s higher price reflects its substantially larger transistor count, more memory, and newer architecture. The GTX 460’s place in the database’s percentile rankings is 41, while the GTX 970 sits at 39, but those rankings are based on different benchmark suites and should not be read as a direct performance ordering. The only direct comparison, OpenCL, shows the GTX 970 ahead by a wide margin.