NVIDIA GeForce GTX 460 v2 ES

The NVIDIA GeForce GTX 460 v2 ES is a 40 nm Fermi 2.0 part built on the GF114 chip at TSMC, packing 1,950 million transistors on a 332 mm² die. Released on 2011-09-23, it sits in the GeForce 400 generation, with a predecessor in GeForce 200 and a successor in GeForce 500. The database records an average benchmark score of 0 and a percentile ranking of 50, placing it exactly at the median of all tracked GPUs.

Benchmark Performance

The benchmark data shows a 0 average score, which is an outlier, but the 50th percentile placement indicates it sits at the median of all GPUs. The raw compute figures tell a clearer story: FP32 performance is 1,046.3 GFLOPS, driven by 336 shading units. The pixel rate of 10.91 GPixel/s and texture rate of 43.62 GTexel/s are derived from 32 ROPs and 56 TMUs respectively. These numbers suggest a mid-range part of its era, but without nearestRivals data, direct percentage comparisons are unavailable. The 1,046.3 GFLOPS figure is the key compute metric, and the 43.62 GTexel/s texture rate indicates how quickly it can feed shaders. The 10.91 GPixel/s pixel rate limits fill-rate-heavy workloads. Since the database lists no rival scores, the analysis must rely on these absolute figures. The 50th percentile is a strong indicator that it outperforms half of all GPUs tracked, but the 0 benchmark score suggests the card may not have been tested under standard conditions, or the score is a placeholder. Nevertheless, the compute metrics are consistent with a 336-shader, 56-TMU, 32-ROP configuration. The transistor density of 5.9M / mm² is a product of 1,950 million transistors spread over 332 mm², which is typical for a 40 nm process. The FP32 throughput of 1,046.3 GFLOPS is the peak single-precision rate, and it is a useful baseline for comparing raw compute capability. In the absence of rival deltas, the 50th percentile is the only comparative metric, and it places this card exactly in the middle of the database's tracked GPUs. This means it is neither a high-end nor a low-end part; it is a true median performer. The 0 benchmark score, however, is problematic; it could indicate a lack of test data for this ES variant, or a failure to complete standard benchmarks. The percentile is based on the full database, so it is more trustworthy than the raw score. Overall, the compute metrics are consistent with a card that would handle 1080p gaming at medium settings in its era, but modern titles would strain its 1,046.3 GFLOPS.

Power and Cooling

The GTX 460 v2 ES has a TDP of 160 W, which is modest for a dual-slot card. The suggested PSU rating is 450 W, and the card requires two 6-pin power connectors. The dual-slot cooler is necessary to dissipate 160 W of heat. The card measures 210 mm (8.3 inches) in length, which fits most mid-tower cases. The 450 W PSU recommendation is a hard requirement; pairing this card with a lower-wattage unit could cause instability. The two 6-pin connectors are standard for this class. The 160 W TDP is a fixed specification, so power draw should not exceed that under load. The dual-slot design means it will occupy two expansion slots. The 210 mm length is a physical constraint for smaller cases. The 2x 6-pin connectors require that the PSU have two dedicated PCIe power cables; adapters are not recommended. The 450 W PSU figure is the suggested minimum, and it accounts for the rest of the system's draw. The 160 W TDP is the thermal design power, which dictates the cooling solution; a dual-slot cooler with a large heatsink is adequate. The 8.3-inch length (210 mm) is a useful measurement for case compatibility. The card's power connectors are 2x 6-pin, which is a common configuration for mid-range GPUs of that generation. The 450 W PSU recommendation is a guideline, but a quality 450 W unit is sufficient. The card's TDP of 160 W is not excessive, so it does not require a high-end power supply.

Memory Subsystem

The memory subsystem is built around 1280 MB of GDDR5 on a 256-bit bus, yielding a bandwidth of 128.3 GB/s. The memory clock is 1002 MHz, which translates to 4 Gbps effective. This configuration provides a balanced amount of bandwidth for the compute power. For high resolutions, the 1280 MB capacity is a limiting factor; modern textures can exceed this easily. The 256-bit bus is a positive, as it allows the 128.3 GB/s bandwidth to be sustained. At 1080p and lower, the 1280 MB is adequate for older titles, but for 1440p or 4K, the capacity becomes a bottleneck. The 128.3 GB/s bandwidth is sufficient for the 1,046.3 GFLOPS compute rate, meaning the card is not severely bandwidth-starved at its intended resolution. The 4 Gbps effective memory speed is typical for GDDR5 of that era. The 1280 MB capacity is unusual; it is not a standard 1 GB or 2 GB, but it is a specific configuration for this ES variant. The 256-bit bus width is a key advantage, as it doubles the bandwidth compared to a 128-bit bus at the same memory clock. The 128.3 GB/s figure is the result of the 256-bit bus multiplied by the 4 Gbps effective speed. The memory subsystem is well-balanced for the GPU's compute capabilities, but the 1280 MB capacity will be the first limitation in high-resolution scenarios. For users who play at 1080p with older games, the 1280 MB is sufficient, but for modern titles with high-resolution textures, it will cause stuttering or texture pop-in. The 1002 MHz memory clock is the base frequency, and the 4 Gbps effective is the data rate.

How It Compares

The database lists no nearest rivals for this GPU, which means direct percentage deltas are not available. Its position at the 50th percentile of all GPUs, however, provides a benchmark-agnostic comparison. The card is part of the GeForce 400 generation, succeeding the GeForce 200 series and preceding the GeForce 500 series. Without rival scores, the comparison must be qualitative. The 50th percentile suggests it sits at the median of the entire GPU landscape, which is a reasonable position for a mid-range part from 2011. The lack of a nearestRivals entry is notable; it implies the database does not have a direct comparison set for this specific ES (engineering sample) variant. The v2 ES designation may indicate a revised version of the original GTX 460, but the database does not provide a baseline for comparison. The 0 average benchmark score further complicates direct comparison, as it prevents any percentage-based analysis. Nevertheless, the 50th percentile is a useful anchor, indicating that half of all GPUs are slower and half are faster. The predecessor, GeForce 200, and successor, GeForce 500, bracket this card in the product stack. The GeForce 400 generation is known for its Fermi architecture, and this card uses the Fermi 2.0 revision. The 50th percentile is a strong statement: it is not a budget part, nor a flagship, but a median performer. The lack of rival data means we cannot say "30% ahead of X" or "20% behind Y," but the percentile gives a global context. The card's release date of 2011-09-23 places it late in the GeForce 400 cycle, and its end-of-life production status means it is no longer manufactured. The 50th percentile is the single most important comparative figure, and it is a reliable indicator of its standing relative to the entire database.

Ray Tracing and Feature Set

The GTX 460 v2 ES has no RT cores and no tensor cores, as these fields are null in the database. This means it does not support hardware ray tracing or tensor-accelerated workloads. The API support includes DirectX 12 (feature level 11_0) and OpenGL 4.6. Vulkan is not listed, so it is not supported. The DirectX 12 (11_0) support is a baseline feature set; it allows the card to run DirectX 12 titles but only at the 11_0 feature level, which lacks many modern features like mesh shaders or variable rate shading. OpenGL 4.6 is a mature API, but the card's compute capabilities are limited by its Fermi 2.0 architecture. The lack of RT and tensor cores means it is not suitable for ray-traced games or AI workloads. The feature set is purely rasterization-based, which aligns with its 2011 release date. The 2x DVI and 1x mini-HDMI 1.3a outputs provide display connectivity, but the lack of DisplayPort is a limitation for modern monitors. The DirectX 12 (11_0) feature level is a subset of the full DirectX 12 API, so some games will run, but with reduced features. OpenGL 4.6 is supported, which is good for legacy applications. The absence of Vulkan support means no Vulkan-based games or applications can run. The RT and tensor core fields are null, confirming that the Fermi 2.0 architecture does not include dedicated ray tracing or tensor processing units. The 2x DVI outputs are dual-link, capable of driving high-resolution monitors, but the 1x mini-HDMI 1.3a is limited to 1080p at 60 Hz. The feature set is a product of its time, and it lacks all modern acceleration features.

Who Should Consider It

The GTX 460 v2 ES is best suited for users targeting lower resolutions and older games. The 1280 MB VRAM and 128.3 GB/s bandwidth are adequate for 1080p and below, but they will be strained at higher resolutions. The 1,046.3 GFLOPS FP32 performance and 43.62 GTexel/s texture rate can handle games from its release era, but modern titles will likely exceed its capabilities. The 50th percentile placement indicates it is a median performer, so it will not excel at demanding workloads. The 160 W TDP and 450 W PSU requirement are reasonable for a system of its age. The card is end-of-life, so it is not a purchase recommendation for new builds. For retro gaming or budget builds using older hardware, the 2x DVI and 1x mini-HDMI outputs are sufficient. The lack of RT and tensor cores means it is purely a rasterization card. The 256-bit bus and 128.3 GB/s bandwidth are its strengths, but the 1280 MB capacity is its primary weakness. If you are building a period-correct system, this card fits; for modern gaming, it is inadequate. The 0 benchmark score suggests that no reliable performance data exists, so users should rely on the 50th percentile and the compute metrics. The 40 nm process and 1,950 million transistors are historical facts, but they do not translate to modern efficiency. The card's 210 mm length (8.3 inches) makes it easy to install in most cases, but the dual-slot cooler will block adjacent slots. For users with a 450 W PSU, the 160 W TDP is within limits, but the 2x 6-pin connectors must be present. The 1280 MB VRAM is the hard ceiling; any game that requires more than that will fail. The 128.3 GB/s bandwidth is sufficient for the compute rate, but it is not future-proof. The card is a product of its generation, and its 50th percentile standing is a testament to its median performance. For a collector or a retro gamer, it is a viable option; for anyone else, it is a historical artifact.