AMD Radeon HD 6990
AMD graphics card specifications and benchmark scores
At a Glance
AMDAMD Radeon HD 6990 Specifications
Radeon HD 6990 GPU Core
Shader units and compute resources
The AMD Radeon HD 6990 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.
HD 6990 Clock Speeds
GPU and memory frequencies
Clock speeds directly impact the Radeon HD 6990'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 Radeon HD 6990 by AMD dynamically adjusts frequencies based on workload, temperature, and power limits to maximize performance while maintaining stability.
AMD's Radeon HD 6990 Memory
VRAM capacity and bandwidth
VRAM (Video RAM) is dedicated memory for storing textures, frame buffers, and shader data. The Radeon HD 6990'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.
Radeon HD 6990 by AMD Cache
On-chip cache hierarchy
On-chip cache provides ultra-fast data access for the HD 6990, 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.
HD 6990 Theoretical Performance
Compute and fill rates
Theoretical performance metrics provide a baseline for comparing the AMD Radeon HD 6990 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.
TeraScale 3 Architecture & Process
Manufacturing and design details
The AMD Radeon HD 6990 is built on AMD's TeraScale 3 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 HD 6990 will perform in GPU benchmarks compared to previous generations.
AMD's Radeon HD 6990 Power & Thermal
TDP and power requirements
Power specifications for the AMD Radeon HD 6990 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 Radeon HD 6990 to maintain boost clocks without throttling.
Radeon HD 6990 by AMD Physical & Connectivity
Dimensions and outputs
Physical dimensions of the AMD Radeon HD 6990 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.
AMD API Support
Graphics and compute APIs
API support determines which games and applications can fully utilize the AMD Radeon HD 6990. 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.
Radeon HD 6990 Product Information
Release and pricing details
The AMD Radeon HD 6990 is manufactured by AMD 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 Radeon HD 6990 by AMD represents good value at current market prices. Predecessor and successor information aids in tracking generational improvements and planning future upgrades.
Radeon HD 6990 Benchmark Scores
No benchmark data available for this GPU.
About AMD Radeon HD 6990
The AMD Radeon HD 6990, released on March 7, 2011, is an end-of-life dual-GPU graphics card built on the TeraScale 3 architecture and the 40 nm process from TSMC. It packs 2,640 million transistors on a 389 mm² die, yielding a transistor density of 6.8 million per square millimeter. The card occupies a dual-slot form factor, measures 310 mm in length, 115 mm in height, and 40 mm in width, and was positioned as a high-end offering in the Northern Islands (HD 6900) generation, sitting between the Evergreen and Southern Islands architectures. With a launch MSRP of 699 USD, it was a premium product at its time, and the database places it at the 50th percentile among all GPUs, indicating it performs at the median of historical hardware.
Memory Subsystem
The HD 6990 is equipped with 2 GB of GDDR5 memory across a 256-bit bus. The memory clock runs at 1250 MHz, translating to an effective data rate of 5 Gbps and a peak bandwidth of 160.0 GB/s. This configuration was typical for a high-end card of its era, providing enough capacity and throughput for demanding titles at high resolutions. The 256-bit bus width strikes a balance between memory capacity and bandwidth efficiency, allowing the card to handle large texture sets without becoming the primary bottleneck. At 160.0 GB/s, the bandwidth supports smooth performance in 1080p and 1440p workloads, though it may struggle with very large frame buffers or extreme anti-aliasing at higher resolutions. The 2 GB capacity is sufficient for most games of the 2011–2014 period, but it can be limiting for modern titles that require more VRAM for ultra-texture packs or high-resolution rendering. For multi-monitor setups, the card's four mini-DisplayPort 1.2 outputs and one DVI port allow flexible display configurations, though the memory subsystem's bandwidth must be shared across all active displays.
Ray Tracing and Feature Set
The HD 6990 does not include dedicated ray tracing cores or tensor cores, as those technologies were not part of the TeraScale 3 architecture. Instead, it relies on traditional rasterization pipelines, with 1536 shading units, 96 texture mapping units, and 32 ROPs. The card supports DirectX 11.2 (with the 11_0 feature level) and OpenGL 4.4, but it lacks Vulkan support, which limits its compatibility with modern graphics APIs that rely on low-level hardware access. For ray tracing, the absence of dedicated hardware means any such effects would have to be computed via software or shader-based approximations, which would be impractical given the card's compute throughput. The feature set is firmly rooted in the DirectX 11 era, making it suitable for games designed around that API but not for titles that require DirectX 12 Ultimate or Vulkan ray tracing. The card's pixel rate of 26.56 GPixel/s and texture rate of 79.68 GTexel/s indicate strong fill-rate capabilities for its generation, which are essential for high-resolution rasterization but do not contribute to ray tracing performance.
Benchmark Performance
The database lists an average benchmark score of 0 for the HD 6990, meaning no validated benchmark runs are recorded in this dataset. Consequently, direct performance comparisons with specific rivals cannot be made from the available data. However, the card's percentile rank of 50 places it exactly at the median of all GPUs in the database, suggesting that its real-world performance is neither exceptional nor deficient relative to the entire historical GPU landscape. The theoretical compute metrics provide additional context: the FP32 throughput is 2.550 TFLOPS, which was competitive for a high-end card in 2011. The pixel rate of 26.56 GPixel/s and texture rate of 79.68 GTexel/s are consistent with a card that can handle 1080p gaming at high settings and 1440p at reduced quality. The 2 GB VRAM and 160.0 GB/s bandwidth are adequate for the era's game requirements, but they would likely bottleneck modern titles that demand more memory bandwidth and capacity. Given the lack of benchmark scores, the percentile rank and theoretical specifications are the primary indicators of performance, and they suggest a mid-tier historical standing rather than a top-tier one.
How It Compares
Without a list of nearest rivals, the HD 6990's positioning must be inferred from its generation and architectural lineage. As a Northern Islands product, it sits between the Evergreen architecture (its predecessor) and Southern Islands (its successor). The Evergreen generation introduced DirectX 11 support and a scalable VLIW architecture, while Southern Islands brought GCN (Graphics Core Next) and improved compute efficiency. The HD 6990's TeraScale 3 design represents the final refinement of the VLIW approach, with higher clock speeds and better power efficiency than its predecessor, though it lacks the compute-oriented features of its successor. The 50th percentile rank indicates that, across all GPUs in the database, it performs at the median—neither a standout nor a laggard. In the context of its own era, the card was a high-end dual-GPU solution, but its 2 GB memory capacity and 256-bit bus were shared between two GPUs, which could lead to frame pacing and driver complexity issues. The absence of Vulkan support further limits its modern relevance, as many contemporary games rely on that API for performance. Overall, the HD 6990 is a historical artifact that occupies a middle ground in the performance spectrum, outperforming integrated graphics and older discrete cards but falling well behind modern mid-range offerings.
Who Should Consider It
The HD 6990 is best suited for users running legacy systems that rely on DirectX 11 titles and do not require modern API features. Its 2 GB VRAM and 160.0 GB/s bandwidth can handle 1080p gaming at medium to high settings for games released before 2015, and its four mini-DisplayPort outputs make it viable for multi-monitor productivity or non-gaming workloads. The card's 2.550 TFLOPS FP32 performance is sufficient for older compute tasks, but it is not recommended for current AAA games, which typically require more than 4 GB of VRAM and support for Vulkan or DirectX 12. Users who want to play modern titles at 1080p with high settings should look for newer hardware, as the HD 6990's memory capacity and bandwidth will likely cause stuttering or texture pop-in. For retro gaming or as a collector's piece, the card offers a glimpse into the dual-GPU era, but its 375 W TDP and 750 W PSU requirement make it an inefficient choice for daily use. The 50th percentile rank reinforces that this is a mid-tier performer by historical standards, not a high-end solution for today's workloads.
FAQ
Q: What is the memory configuration of the AMD Radeon HD 6990?
A: The card features 2 GB of GDDR5 memory on a 256-bit bus, with a memory clock of 1250 MHz (5 Gbps effective) and a peak bandwidth of 160.0 GB/s.
Q: Does the HD 6990 support ray tracing or tensor cores?
A: No. The card has no dedicated ray tracing or tensor cores, and its architecture (TeraScale 3) relies on traditional rasterization. It supports DirectX 11.2 and OpenGL 4.4, but lacks Vulkan support.
Q: What is the recommended power supply for this card?
A: The suggested PSU is 750 W, and the card requires two 8-pin power connectors. Its TDP is 375 W.
Q: What is the process node and transistor count?
A: The HD 6990 is fabricated on a 40 nm process at TSMC, containing 2,640 million transistors on a 389 mm² die.
Q: How many display outputs does it have?
A: It has one DVI port and four mini-DisplayPort 1.2 outputs, supporting multi-monitor setups.
Q: What is the card's performance percentile?
A: The database places it at the 50th percentile among all GPUs, indicating it performs at the median of historical hardware. Its average benchmark score is 0, meaning no validated benchmark runs are recorded.
Power and Cooling
The HD 6990 has a TDP of 375 W, making it a power-hungry card that demands a robust power supply. AMD recommends a 750 W PSU, and the card requires two 8-pin power connectors for operation. The dual-slot cooler is designed to dissipate the heat generated by the two GPUs, but the card's 310 mm length and 115 mm height require a spacious case with adequate airflow. The 40 mm width means it occupies two expansion slots, which can limit adjacent PCIe slots. The 40 nm process and 2,640 million transistors contribute to the high power draw, and the card's end-of-life status means no further driver optimizations are likely. For users considering this card today, the power and cooling requirements are substantial, and the 375 W TDP alone exceeds the total power budget of many modern mid-range systems. The 2x 8-pin connectors are standard for high-end cards of that era, but they may not be available on older power supplies. Overall, the power and cooling demands are a significant consideration, and the card's efficiency is poor by modern standards, but it remains a functional piece of hardware for legacy builds that can accommodate its physical and electrical needs.
The NVIDIA Equivalent of Radeon HD 6990
Looking for a similar graphics card from NVIDIA? The NVIDIA GeForce RTX 2080 offers comparable performance and features in the NVIDIA lineup.
Popular AMD Radeon HD 6990 Comparisons
See how the Radeon HD 6990 stacks up against similar graphics cards from the same generation and competing brands.
Compare Radeon HD 6990 with Other GPUs
Select another GPU to compare specifications and benchmarks side-by-side.
Browse GPUs