Tesla K40t vs Radeon HD 8670D
Contents
Primary details
GPU architecture, market segment, value for money and other general parameters compared.
| Place in the ranking | 1107 | not rated |
| Place by popularity | not in top-100 | not in top-100 |
| Power efficiency | 0.94 | no data |
| Architecture | TeraScale 3 (2010−2013) | Kepler (2012−2018) |
| GPU code name | Devastator | GK110B |
| Market segment | Desktop | Workstation |
| Release date | 12 March 2013 (13 years ago) | 22 November 2013 (12 years ago) |
| Launch price (MSRP) | no data | $7,699 |
Detailed specifications
General parameters such as number of shaders, GPU core base clock and boost clock speeds, manufacturing process, texturing and calculation speed. Real power consumption of some graphics cards can well exceed their nominal TDP, especially if overclocked.
| Pipelines / CUDA cores | 384 | 2880 |
| Core clock speed | 844 MHz | 745 MHz |
| Boost clock speed | 950 MHz | 876 MHz |
| Number of transistors | 1,303 million | 7,080 million |
| Manufacturing process technology | 32 nm | 28 nm |
| Power consumption (TDP) | 100 Watt | 245 Watt |
| Texture fill rate | 22.80 | 210.2 |
| Floating-point processing power | 0.7296 TFLOPS | 5.046 TFLOPS |
| ROPs | 8 | 48 |
| TMUs | 24 | 240 |
| L1 Cache | no data | 240 KB |
| L2 Cache | no data | 1536 KB |
Form factor & compatibility
Information on compatibility with other computer components. Useful when choosing a future computer configuration or upgrading an existing one. For desktop graphics cards it's interface and bus (motherboard compatibility), additional power connectors (power supply compatibility).
| Interface | IGP | PCIe 3.0 x16 |
| Length | no data | 267 mm |
| Width | IGP | 2-slot |
VRAM capacity and type
Parameters of VRAM installed: its type, size, bus, clock and resulting bandwidth. Integrated GPUs have no dedicated video RAM and use a shared part of system RAM.
| Memory type | System Shared | GDDR5 |
| Maximum RAM amount | System Shared | 12 GB |
| Memory bus width | System Shared | 384 Bit |
| Memory clock speed | System Shared | 1502 MHz |
| Memory bandwidth | no data | 288.4 GB/s |
| Shared memory | + | - |
Connectivity and outputs
This section shows the types and number of video connectors on each GPU. The data applies specifically to desktop reference models (for example, NVIDIA’s Founders Edition). OEM partners often modify both the number and types of ports. On notebook GPUs, video‐output options are determined by the laptop’s design rather than the graphics chip itself.
| Display Connectors | No outputs | No outputs |
API and SDK support
List of supported 3D and general-purpose computing APIs, including their specific versions.
| DirectX | 11.2 (11_0) | 12 (11_1) |
| Shader Model | 5.0 | 5.1 |
| OpenGL | 4.4 | 4.6 |
| OpenCL | 1.2 | 1.2 |
| Vulkan | N/A | 1.1.126 |
| CUDA | - | 3.5 |
Pros & cons summary
| Recency | 12 March 2013 | 22 November 2013 |
| Chip lithography | 32 nm | 28 nm |
| Power consumption (TDP) | 100 Watt | 245 Watt |
HD 8670D has 145% lower power consumption.
Tesla K40t, on the other hand, has an age advantage of 8 months, and a 14% more advanced lithography process.
We couldn't decide between Radeon HD 8670D and Tesla K40t. We've got no test results to judge.
Be aware that Radeon HD 8670D is a desktop graphics card while Tesla K40t is a workstation one.
Other comparisons
We selected several comparisons of graphics cards with performance close to those reviewed, providing you with more options to consider.
