NVIDIA GeForce GTX 680
NVIDIA graphics card specifications and benchmark scores
At a Glance
NVIDIANVIDIA GeForce GTX 680 Specifications
GeForce GTX 680 GPU Core
Shader units and compute resources
The NVIDIA GeForce GTX 680 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 680 Clock Speeds
GPU and memory frequencies
Clock speeds directly impact the GeForce GTX 680'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 680 by NVIDIA dynamically adjusts frequencies based on workload, temperature, and power limits to maximize performance while maintaining stability.
NVIDIA's GeForce GTX 680 Memory
VRAM capacity and bandwidth
VRAM (Video RAM) is dedicated memory for storing textures, frame buffers, and shader data. The GeForce GTX 680'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 680 by NVIDIA Cache
On-chip cache hierarchy
On-chip cache provides ultra-fast data access for the GTX 680, 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 680 Theoretical Performance
Compute and fill rates
Theoretical performance metrics provide a baseline for comparing the NVIDIA GeForce GTX 680 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 680 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 680 will perform in GPU benchmarks compared to previous generations.
NVIDIA's GeForce GTX 680 Power & Thermal
TDP and power requirements
Power specifications for the NVIDIA GeForce GTX 680 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 680 to maintain boost clocks without throttling.
GeForce GTX 680 by NVIDIA Physical & Connectivity
Dimensions and outputs
Physical dimensions of the NVIDIA GeForce GTX 680 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 680. 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 680 Product Information
Release and pricing details
The NVIDIA GeForce GTX 680 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 680 by NVIDIA represents good value at current market prices. Predecessor and successor information aids in tracking generational improvements and planning future upgrades.
GeForce GTX 680 Benchmark Scores
geekbench_metalSource
Geekbench Metal tests GPU compute using Apple's Metal API. This shows how NVIDIA GeForce GTX 680 performs in macOS and iOS applications that leverage GPU acceleration. Metal provides low-overhead access to Apple silicon GPUs.
Top 5 Performers
Top 5 PerformersNearby Performers
Nearby Performersgeekbench_openclSource
Geekbench OpenCL tests GPU compute performance using the cross-platform OpenCL API. This shows how NVIDIA GeForce GTX 680 handles parallel computing tasks like video encoding and scientific simulations.
Top 5 Performers
Nearby Performers
geekbench_vulkanSource
Geekbench Vulkan tests GPU compute using the modern low-overhead Vulkan API. This shows how NVIDIA GeForce GTX 680 performs with next-generation graphics and compute workloads. Vulkan offers better CPU efficiency than older APIs like OpenGL.
Top 5 Performers
Nearby Performers
About NVIDIA GeForce GTX 680
Launched in March 2012 on TSMC’s 28 nm process, the NVIDIA GeForce GTX 680 introduced the Kepler architecture with 3,540 million transistors packed into a 294 mm² die. It carries 1,536 shading units, 128 texture mapping units, and 32 ROPs, with a base clock of 1006 MHz and a boost clock of 1058 MHz. The card delivers 3.250 TFLOPS of FP32 compute, a pixel rate of 33.86 GPixel/s, and a texture rate of 135.4 GTexel/s. It was positioned as an end-of-life product, succeeding the GeForce 500 series and preceding the GeForce 700 series, with a launch MSRP of 499 USD.
Benchmark Performance
The GTX 680’s average benchmark score stands at 13,812, placing it in the 54th percentile of all GPUs. This is a mid-pack ranking, indicating that while the card is no longer competitive at the top end, it remains functional for a range of applications. The data shows a tight cluster of rivals around this score, with the GTX 680 effectively tied with the AMD Radeon 660M, which also scores 13,812, representing a 0% delta. This suggests that, in aggregate compute and graphics workloads, the Kepler-based card matches a modern integrated graphics solution, which is a notable outcome given the architectural and process node differences.
In individual API tests, the GTX 680 shows significant variance. Its Geekbench OpenCL score of 16,803 is substantially higher than its Metal score of 6,987, a gap of roughly 140%. This indicates that the card’s compute capabilities are heavily dependent on the API and driver optimizations, with OpenCL workloads extracting far more performance than Metal. The Vulkan score of 17,646 is the highest of the three, edging out OpenCL by about 5%, suggesting that the card’s asynchronous compute and draw call handling are better suited to Vulkan’s lower-overhead model.
Relative to its nearest rivals, the GTX 680 trails the NVIDIA RTX A2000 Mobile by a negligible 0.1% (13,821 vs. 13,812), a difference well within run-to-run variance. It also sits 0.4% behind the AMD Radeon RX 570X, which scores 13,871. Conversely, it leads the NVIDIA P106-090 by 0.7% (13,716 vs. 13,812). These deltas are all sub-1%, meaning that from a purely synthetic benchmark perspective, the GTX 680, the Radeon 660M, the RTX A2000 Mobile, the RX 570X, and the P106-090 are functionally interchangeable in average score. The real differentiators would come from driver maturity, feature support, and sustained clock behavior, none of which are captured in these aggregate numbers.
The 54th percentile ranking is telling: it means that roughly 46% of all GPUs in the database outperform the GTX 680, while 54% are slower. For a card that was once a flagship, this places it firmly in the lower half of current hardware, but not at the very bottom. The data suggests that the GTX 680 retains enough compute headroom for light workloads, but it has been decisively overtaken by both modern integrated graphics and entry-level discrete solutions.
Who Should Consider It
Benchmark results indicate that the GTX 680 is suited for 1080p gaming at medium to low settings in modern titles, but it will struggle with high refresh rates or high-detail presets. The OpenCL score of 16,803 suggests that compute-heavy applications like video encoding or physics simulations may run acceptably, but the Metal score of 6,987 points to poor performance in macOS-centric workflows that rely on Metal acceleration. Users who primarily play older games, esports titles, or indie games from the early 2010s will find the card serviceable, especially at 1080p with reduced shadows and anti-aliasing.
At 1440p, the data does not support a recommendation. The pixel rate of 33.86 GPixel/s and texture rate of 135.4 GTexel/s are too low to maintain smooth frame rates in demanding scenes. The card’s 32 ROPs are a bottleneck for high-resolution fill-rate workloads, and the 192.3 GB/s bandwidth will limit texture streaming and memory-intensive effects. For users with a 1440p monitor, the GTX 680 is best left for desktop use or non-gaming tasks. At 4K, the card is effectively unusable for gaming, as the fill rate and memory bandwidth are insufficient to push the pixel count.
The 450 W suggested PSU and dual 6-pin power connectors indicate that this card can be installed in older systems with adequate power supplies, but its 195 W TDP means it will generate substantial heat. The dual-slot cooler is standard for the era, but users in small form factor cases should verify clearance given the 256 mm length. The card’s DirectX 12 (11_0) support is a caveat: while it can run DX12 titles, it is limited to the 11_0 feature level, which means it will miss out on advanced features like mesh shaders or variable rate shading. This makes it a poor choice for users aiming to play the latest AAA releases, even at low settings.
Memory Subsystem
The GTX 680 is equipped with 2 GB of GDDR5 memory on a 256-bit bus, yielding a bandwidth of 192.3 GB/s. The memory clock is 1502 MHz, with an effective data rate of 6 Gbps. For a 2012 flagship, this was a competitive configuration, but by modern standards, 2 GB is a hard limitation. At 1080p, many current games require more than 2 GB for high-resolution textures, and the card will either drop textures to lower mip levels or stutter due to memory swapping. The 256-bit bus width provides decent throughput for the era, but the bandwidth is roughly half of what mid-range cards deliver today.
At high resolutions like 1440p or 4K, the memory subsystem becomes the primary bottleneck. The 192.3 GB/s bandwidth is insufficient to feed the 32 ROPs and 1,536 shading units at those pixel counts, leading to severe frame time spikes. The 2 GB capacity also means that frame buffers for high-resolution textures will exceed the available memory, causing significant performance degradation. In benchmarks, the card’s average score of 13,812 reflects this limitation; it is not a compute-bound failure but rather a memory-bound one.
The 6 Gbps effective memory speed is moderate, and the card does not benefit from any memory compression techniques that are common in newer architectures. The data suggests that for any workload that exceeds 2 GB of memory usage, the GTX 680 will underperform its raw compute potential. Users should treat the card as a 1080p-only solution with texture quality set to medium or high, but never ultra, and should avoid any resolution scaling beyond 1080p.
FAQ
Q: What is the average benchmark score of the GTX 680?
A: The average benchmark score is 13,812, which places it in the 54th percentile of all GPUs.
Q: How does the GTX 680 compare to the AMD Radeon 660M?
A: They are exactly tied, with both scoring 13,812, representing a 0% delta in average benchmark performance.
Q: What is the card’s memory bandwidth and bus width?
A: It has 2 GB of GDDR5 memory on a 256-bit bus, providing 192.3 GB/s of bandwidth.
Q: Does the GTX 680 support DirectX 12 Ultimate features?
A: No, it supports DirectX 12 at the 11_0 feature level, meaning it lacks advanced DX12 features like mesh shaders and variable rate shading.
Q: What is the launch MSRP of the GTX 680?
A: The launch MSRP was 499 USD.
Q: How does the GTX 680 perform in Metal versus Vulkan benchmarks?
A: It scores 6,987 in Geekbench Metal and 17,646 in Geekbench Vulkan, showing a 152% advantage for Vulkan.
How It Compares
vs. AMD Radeon 660M: The GTX 680 and the Radeon 660M are statistically identical, with a 0% delta in average score (13,812 for both). This is surprising because the 660M is a modern integrated GPU, while the GTX 680 is a discrete card from 2012. The data indicates that for synthetic benchmarks, the two are interchangeable, but the 660M likely draws far less power and supports newer APIs, making it a more practical choice in modern systems.
vs. NVIDIA RTX A2000 Mobile: The RTX A2000 Mobile edges out the GTX 680 by a mere 0.1%, scoring 13,821 versus 13,812. This sub-1% difference means that in real-world workloads, the two cards would perform almost identically in compute tasks. However, the A2000 Mobile is a laptop part with likely lower TDP and better driver support for professional applications, whereas the GTX 680 is a desktop card with older architecture.
vs. AMD Radeon RX 570X: The RX 570X leads by 0.4%, with a score of 13,871 versus 13,812. This is a margin that would not be noticeable in gaming or content creation. The RX 570X is a newer card with likely better power efficiency and more modern feature support, but the benchmark data shows that the GTX 680 remains competitive in raw compute throughput.
vs. NVIDIA P106-090: The GTX 680 is ahead of the P106-090 by 0.7%, scoring 13,812 versus 13,716. The P106-090 is a mining-oriented card with no display outputs, so the comparison is purely academic. The GTX 680’s slight lead in average score suggests that its Kepler architecture has a marginal compute advantage over the P106-090’s Pascal-based design in these specific workloads.
The AMD Equivalent of GeForce GTX 680
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 680 Comparisons
See how the GeForce GTX 680 stacks up against similar graphics cards from the same generation and competing brands.
Compare GeForce GTX 680 with Other GPUs
Select another GPU to compare specifications and benchmarks side-by-side.
Browse GPUs