NVIDIA GeForce GTX 660 OEM
NVIDIA graphics card specifications and benchmark scores
At a Glance
NVIDIANVIDIA GeForce GTX 660 OEM Specifications
GeForce GTX 660 OEM GPU Core
Shader units and compute resources
The NVIDIA GeForce GTX 660 OEM GPU core specifications define its raw processing power for graphics and compute workloads. Shading units (also called CUDA cores, stream processors, or execution units depending on manufacturer) handle the parallel calculations required for rendering. TMUs (Texture Mapping Units) process texture data, while ROPs (Render Output Units) handle final pixel output. Higher shader counts generally translate to better GPU benchmark performance, especially in demanding games and 3D applications.
GTX 660 OEM Clock Speeds
GPU and memory frequencies
Clock speeds directly impact the GeForce GTX 660 OEM's performance in GPU benchmarks and real-world gaming. The base clock represents the minimum guaranteed frequency, while the boost clock indicates peak performance under optimal thermal conditions. Memory clock speed affects texture loading and frame buffer operations. The GeForce GTX 660 OEM by NVIDIA dynamically adjusts frequencies based on workload, temperature, and power limits to maximize performance while maintaining stability.
NVIDIA's GeForce GTX 660 OEM Memory
VRAM capacity and bandwidth
VRAM (Video RAM) is dedicated memory for storing textures, frame buffers, and shader data. The GeForce GTX 660 OEM's memory capacity determines how well it handles high-resolution textures and multiple displays. Memory bandwidth, measured in GB/s, affects how quickly data moves between the GPU and VRAM. Higher bandwidth improves performance in memory-intensive scenarios like 4K gaming. The memory bus width and type (GDDR6, GDDR6X, HBM) significantly influence overall GPU benchmark scores.
GeForce GTX 660 OEM by NVIDIA Cache
On-chip cache hierarchy
On-chip cache provides ultra-fast data access for the GTX 660 OEM, reducing the need to fetch data from slower VRAM. L1 and L2 caches store frequently accessed data close to the compute units. AMD's Infinity Cache (L3) dramatically increases effective bandwidth, improving GPU benchmark performance without requiring wider memory buses. Larger cache sizes help maintain high frame rates in memory-bound scenarios and reduce power consumption by minimizing VRAM accesses.
GTX 660 OEM Theoretical Performance
Compute and fill rates
Theoretical performance metrics provide a baseline for comparing the NVIDIA GeForce GTX 660 OEM against other graphics cards. FP32 (single-precision) performance, measured in TFLOPS, indicates compute capability for gaming and general GPU workloads. FP64 (double-precision) matters for scientific computing. Pixel and texture fill rates determine how quickly the GPU can render complex scenes. While real-world GPU benchmark results depend on many factors, these specifications help predict relative performance levels.
Kepler Architecture & Process
Manufacturing and design details
The NVIDIA GeForce GTX 660 OEM is built on NVIDIA's Kepler architecture, which defines how the GPU processes graphics and compute workloads. The manufacturing process node affects power efficiency, thermal characteristics, and maximum clock speeds. Smaller process nodes pack more transistors into the same die area, enabling higher performance per watt. Understanding the architecture helps predict how the GTX 660 OEM will perform in GPU benchmarks compared to previous generations.
NVIDIA's GeForce GTX 660 OEM Power & Thermal
TDP and power requirements
Power specifications for the NVIDIA GeForce GTX 660 OEM determine PSU requirements and thermal management needs. TDP (Thermal Design Power) indicates the heat output under typical loads, guiding cooler selection. Power connector requirements ensure adequate power delivery for stable operation during demanding GPU benchmarks. The suggested PSU wattage accounts for the entire system, not just the graphics card. Efficient power delivery enables the GeForce GTX 660 OEM to maintain boost clocks without throttling.
GeForce GTX 660 OEM by NVIDIA Physical & Connectivity
Dimensions and outputs
Physical dimensions of the NVIDIA GeForce GTX 660 OEM are critical for case compatibility. Card length, height, and slot width determine whether it fits in your chassis. The PCIe interface version affects bandwidth for communication with the CPU. Display outputs define monitor connectivity options, with modern cards supporting multiple high-resolution displays simultaneously. Verify these specifications against your case and motherboard before purchasing to ensure a proper fit.
NVIDIA API Support
Graphics and compute APIs
API support determines which games and applications can fully utilize the NVIDIA GeForce GTX 660 OEM. DirectX 12 Ultimate enables advanced features like ray tracing and variable rate shading. Vulkan provides cross-platform graphics capabilities with low-level hardware access. OpenGL remains important for professional applications and older games. CUDA (NVIDIA) and OpenCL enable GPU compute for video editing, 3D rendering, and scientific applications. Higher API versions unlock newer graphical features in GPU benchmarks and games.
GeForce GTX 660 OEM Product Information
Release and pricing details
The NVIDIA GeForce GTX 660 OEM is manufactured by NVIDIA as part of their graphics card lineup. Release date and launch pricing provide context for comparing GPU benchmark results with competing products from the same era. Understanding the product lifecycle helps evaluate whether the GeForce GTX 660 OEM by NVIDIA represents good value at current market prices. Predecessor and successor information aids in tracking generational improvements and planning future upgrades.
GeForce GTX 660 OEM Benchmark Scores
No benchmark data available for this GPU.
About NVIDIA GeForce GTX 660 OEM
Benchmark Performance
The NVIDIA GeForce GTX 660 OEM is a 2012-era graphics card built on the Kepler architecture, specifically utilizing the GK104 chip. With a transistor count of 3,540 million on a 294 mm² die produced on TSMC's 28 nm process, it represents a mature mid-range design from the GeForce 600 generation. Its performance percentile of 50 against all GPUs places it squarely in the middle of the pack historically, meaning it can still handle a surprising amount of legacy content but is far from competitive with modern hardware.
The card's raw compute capabilities are defined by its 1,152 shading units, 96 texture mapping units, and 24 ROPs. This configuration yields a pixel rate of 21.31 GPixel/s and a texture rate of 85.25 GTexel/s. In practical terms, these figures translate to a card that was designed for 1080p gaming in its era, and the data suggests it still delivers acceptable frame rates in older titles at that resolution. The FP32 performance of 2.046 TFLOPS is modest by today's standards but was respectable for a mid-range Kepler part; this places it well below entry-level modern GPUs, which often exceed 10 TFLOPS, but it remains functional for less demanding workloads.
Because the benchmark data for this specific OEM variant is sparse—the benchmark score is listed as 0 and there are no nearestRivals provided—we must rely on architectural characteristics to contextualize its standing. The GK104 chip is the same silicon used in higher-tier Kepler cards, but the GTX 660 OEM is cut down significantly. The 2.046 TFLOPS figure suggests it operates at a fraction of the throughput of the full GK104 implementation. In a comparative sense, cards from the GeForce 700 series, its direct successor, would typically show a 15-30% improvement in raw shader throughput at similar price points, but without specific rival data, we can only infer that the GTX 660 OEM sits firmly in the lower-middle tier of its generation.
The clock speeds of 823 MHz base and 888 MHz boost are conservatively set, which is typical for OEM parts that prioritize stability over maximum performance. The boost behavior on Kepler is dynamic, but the 65 MHz differential between base and boost indicates limited thermal headroom in the reference design. This card will not win any performance awards today, but for a system running Windows 7-era games or light esports titles, the data indicates it remains serviceable.
Ray Tracing and Feature Set
The GTX 660 OEM does not include dedicated ray tracing cores or tensor cores. This is a fundamental architectural limitation of the Kepler generation—ray tracing acceleration did not appear in NVIDIA's consumer lineup until much later. Consequently, any ray-traced workloads are entirely out of the question on this hardware; the card lacks the fixed-function hardware to accelerate BVH traversal and ray intersection calculations.
What the card does support is a feature set appropriate for its time. DirectX 12 (11_0) support is present, though the 11_0 feature level means it cannot take advantage of DirectX 12 Ultimate features like mesh shaders or variable rate shading. OpenGL 4.6 and Vulkan 1.2.175 are both supported, which is better than many competitors from the same era. This means the card can run modern Vulkan-based titles in a compatibility mode, though performance will be limited by the older shader architecture. For practical purposes, a user should treat this card as a DirectX 11-era device; the Vulkan support is a bonus for older open-source drivers or lightweight game engines.
The display outputs include 2x DVI, 1x HDMI 1.4a, and 1x DisplayPort 1.2. The HDMI 1.4a standard caps 4K output at 30 Hz, which is limiting for modern displays. DisplayPort 1.2 does allow 4K at 60 Hz, but the card's memory subsystem (detailed below) would struggle to drive such resolutions in gaming. The lack of any modern connectivity like HDMI 2.1 or DisplayPort 1.4a means this card is best paired with 1080p monitors or older 1440p panels.
Memory Subsystem
The GTX 660 OEM comes equipped with 1536 MB of GDDR5 memory on a 192-bit bus. This is an unusual configuration—most cards in this class used 2 GB or 1 GB—and it reflects the OEM-specific nature of this part. The memory clock is 1400 MHz, which translates to 5.6 Gbps effective, yielding a total bandwidth of 134.4 GB/s. This bandwidth figure is the single most important spec for understanding the card's behavior at higher resolutions.
At 1080p, 134.4 GB/s is adequate for the era's games, which typically required 20-30 GB/s for smooth operation. However, the 1536 MB capacity is a bottleneck even in older titles. Many games from 2013-2015 allocated textures based on available VRAM, and 1.5 GB is right at the threshold where texture pop-in and stuttering become noticeable. The 192-bit bus width is also narrower than the 256-bit buses found on competing mid-range cards, which means the card relies on higher memory clocks to compensate. The result is a memory subsystem that performs well at 1080p with medium textures but degrades quickly if you push settings higher or attempt 1440p.
For 1440p gaming, the 134.4 GB/s bandwidth becomes a serious constraint. Modern titles at that resolution can require 100+ GB/s just for minimal settings, and the small VRAM pool will cause frequent asset swaps from system memory. The data strongly indicates this card is a 1080p-only solution. Even at 1080p, users should expect to run games at medium settings with high textures only in titles designed for the GTX 600 era. The 24 ROPs also limit fill-rate intensive effects like heavy anti-aliasing, which will further strain the memory bandwidth.
Who Should Consider It
Given the performance percentile of 50, this card sits in the middle of all GPUs ever released. That statistic is misleading, however, because it includes many older and weaker cards. For a modern user, the GTX 660 OEM is only viable for specific scenarios. If you are building a retro gaming PC for Windows XP/Vista/7 titles, this card is a reasonable choice. Games from 2008-2013 will run well at 1080p with high settings, and the Kepler architecture has excellent driver support for legacy DirectX 9/10/11 titles.
For esports titles like CS:GO, League of Legends, or Dota 2, the card can handle these at 1080p with medium-to-high settings, provided you accept that frame rates will vary. The 2.046 TFLOPS FP32 performance is enough for these lighter workloads. However, for any modern AAA game released after 2016, the card is not recommended. The 1536 MB VRAM is below the 4 GB minimum that most current games expect, and the lack of modern feature support means you will miss out on even basic optimizations.
The card is end-of-life, with production status confirmed as such. This means no new drivers are being developed for it, and long-term support is limited to legacy branches. If you already own this card, it can serve as a backup or for a secondary machine. If you are purchasing it second-hand, the data suggests you should only do so at a very low price, as its performance is now equivalent to integrated graphics in many modern CPUs.
FAQ
Q: Can this card run modern games at 1080p?
A: The data indicates this is not viable for post-2016 titles. The 1536 MB VRAM and 134.4 GB/s bandwidth are insufficient for modern texture requirements, and the 2.046 TFLOPS FP32 performance is below the threshold for smooth frame rates in demanding titles.
Q: Does the GTX 660 OEM support ray tracing?
A: No. The card has no ray tracing cores or tensor cores, as it is based on the Kepler architecture from 2012. Ray tracing hardware was not introduced until later generations.
Q: What DirectX version does this card support?
A: It supports DirectX 12 (11_0). This means it meets the minimum feature level for DirectX 12 but cannot access higher-tier features like mesh shaders or ray tracing.
Q: Is 1536 MB of VRAM enough for 1440p gaming?
A: No. The 1536 MB capacity is a bottleneck even at 1080p in modern titles. At 1440p, the card will run out of memory quickly, causing severe stuttering and asset pop-in.
Q: What is the maximum display resolution supported?
A: Through DisplayPort 1.2, the card can output 4K at 60 Hz. However, gaming at that resolution is not practical due to the memory and compute limitations. HDMI 1.4a caps at 4K 30 Hz.
Q: How long is this card?
A: The card measures 241 mm, or 9.5 inches, in length. This makes it compatible with most mid-tower cases, but check clearance for large front fans or drive cages.
Power and Cooling
The GTX 660 OEM has a TDP of 130 W. This is a moderate power draw for its era, but it is not insignificant. The suggested PSU is 300 W, which is a low bar by modern standards. Any quality 300 W power supply from a reputable brand will suffice, but note that this recommendation assumes a typical system with a single hard drive and a mid-range CPU. If you are running a high-end CPU or multiple drives, you should consider a 400 W unit for headroom, though the data does not specify this.
The card requires a single 6-pin power connector. This is standard for the time and is available on virtually all power supplies. The dual-slot design means it occupies two expansion slots, which is expected for a card with a 130 W TDP. The cooling solution is not specified in the data, but the 65 MHz boost differential suggests the reference cooler is adequate for the 888 MHz boost clock.
The 28 nm process node from TSMC is relatively inefficient compared to modern 5 nm or 7 nm parts, meaning the 130 W TDP produces more heat per watt than newer cards. In a well-ventilated case, this is not a concern, but in a small form factor system, the card may run hot and throttle below the 888 MHz boost clock. The PCIe 3.0 x16 interface is backward compatible with older slots, but on a modern system, the card will run at PCIe 3.0 speeds, which is sufficient for its bandwidth needs. The 134.4 GB/s memory bandwidth is well within the PCIe 3.0 x16 limit, so there is no bottleneck from the interface.
For cooling, the dual-slot design with a blower-style cooler (typical for OEM parts) exhausts hot air out of the case, which is beneficial for system thermals. The 241 mm length is compact, fitting in most cases without issue. The card's end-of-life status means replacement coolers are hard to find, so if the fan fails, the card is effectively dead. In summary, the power and cooling requirements are modest, making this an easy card to install in any system with a 300 W PSU and a free 6-pin connector.
The AMD Equivalent of GeForce GTX 660 OEM
Looking for a similar graphics card from AMD? The AMD Radeon RX 480 offers comparable performance and features in the AMD lineup.
Popular NVIDIA GeForce GTX 660 OEM Comparisons
See how the GeForce GTX 660 OEM stacks up against similar graphics cards from the same generation and competing brands.
Compare GeForce GTX 660 OEM with Other GPUs
Select another GPU to compare specifications and benchmarks side-by-side.
Browse GPUs