L40 CNX vs Tesla K20c
Contents
Primary details
GPU architecture, market segment, value for money and other general parameters compared.
| Place in the ranking | 543 | not rated |
| Place by popularity | not in top-100 | not in top-100 |
| Cost-effectiveness evaluation | 0.15 | no data |
| Power efficiency | 2.79 | no data |
| Architecture | Kepler (2012−2018) | Ada Lovelace (2022−2024) |
| GPU code name | GK110 | AD102 |
| Market segment | Workstation | Workstation |
| Release date | 12 November 2012 (12 years ago) | 13 October 2022 (2 years ago) |
| Launch price (MSRP) | $3,199 | no data |
Cost-effectiveness evaluation
The higher the ratio, the better. We use the manufacturer's recommended prices.
Performance to price scatter graph
Detailed specifications
General parameters such as number of shaders, GPU core base clock and boost clock speeds, manufacturing process, texturing and calculation speed. Note that power consumption of some graphics cards can well exceed their nominal TDP, especially when overclocked.
| Pipelines / CUDA cores | 2496 | 18176 |
| Core clock speed | 706 MHz | 1005 MHz |
| Boost clock speed | no data | 2475 MHz |
| Number of transistors | 7,080 million | 76,300 million |
| Manufacturing process technology | 28 nm | 4 nm |
| Power consumption (TDP) | 225 Watt | 300 Watt |
| Texture fill rate | 146.8 | 1,406 |
| Floating-point processing power | 3.524 TFLOPS | 89.97 TFLOPS |
| ROPs | 40 | 192 |
| TMUs | 208 | 568 |
| Tensor Cores | no data | 568 |
| Ray Tracing Cores | no data | 142 |
| L1 Cache | 208 KB | 17.8 MB |
| L2 Cache | 1280 KB | 48 MB |
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 | PCIe 2.0 x16 | PCIe 4.0 x16 |
| Length | 267 mm | 267 mm |
| Width | 2-slot | 2-slot |
| Supplementary power connectors | 1x 6-pin + 1x 8-pin | 1x 16-pin |
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 | GDDR5 | GDDR6 |
| Maximum RAM amount | 5 GB | 24 GB |
| Memory bus width | 320 Bit | 384 Bit |
| Memory clock speed | 1300 MHz | 2250 MHz |
| Memory bandwidth | 208.0 GB/s | 864.0 GB/s |
| Resizable BAR | - | + |
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 | 1x HDMI 2.1, 3x DisplayPort 1.4a |
| HDMI | - | + |
API and SDK support
List of supported 3D and general-purpose computing APIs, including their specific versions.
| DirectX | 12 (11_0) | 12 Ultimate (12_2) |
| Shader Model | 5.1 | 6.7 |
| OpenGL | 4.6 | 4.6 |
| OpenCL | 1.2 | 3.0 |
| Vulkan | 1.1.126 | 1.3 |
| CUDA | 3.5 | 8.9 |
| DLSS | - | + |
Pros & cons summary
| Recency | 12 November 2012 | 13 October 2022 |
| Maximum RAM amount | 5 GB | 24 GB |
| Chip lithography | 28 nm | 4 nm |
| Power consumption (TDP) | 225 Watt | 300 Watt |
Tesla K20c has 33.3% lower power consumption.
L40 CNX, on the other hand, has an age advantage of 9 years, a 380% higher maximum VRAM amount, and a 600% more advanced lithography process.
We couldn't decide between Tesla K20c and L40 CNX. We've got no test results to judge.
Other comparisons
We selected several comparisons of graphics cards with performance close to those reviewed, providing you with more options to consider.
