NVIDIA GeForce RTX 4090 vs NVIDIA Tesla K40c
| |
| Price | $1,599.00 $7,699.00 |
| Our Rating | |
| Brand | NVIDIA NVIDIA |
| Category | GPU Benchmarks GPU Benchmarks |
BASIC GPU INFO
| GPU name | AD102 GK180 |
| Manufacturer | NVIDIA TSMC |
| Released date | Sep 20th, 2022 Oct 8th, 2013 |
| Launch price | $1,599 $7,699 |
| Foundry | TSMC |
| Purpose | gaming |
| GPU codename | AD102-300-A1 |
| Architecture | Ada Lovelace Kepler |
| Generation | GeForce 40 Tesla Kepler (Kxx) |
| Production | Available End-of-life |
| Successor | GeForce 50 Tesla Maxwell |
| Predecessor | GeForce 30 |
BOARD DESIGN
| TDP | 450 W 245 W |
| Slot width | Triple-slot Dual-slot |
| Power connectors | 1x 16-pin 1x 6-pin + 1x 8-pin |
| Length | 304 mm 12 inches N/A |
| Height | 61 mm 2.4 inches N/A |
| Width | 137 mm 5.4 inches N/A |
| Outputs | 1x HDMI 2.1 and 3x DisplayPort 1.4a No outputs available on this GPU |
| Board number | PG139 SKU 330 P2081 SKU 206 |
| Weight | N/A |
| PSU | 850 W |
| Bus interface | PCIe 4.0 x16 |
GRAPHICS PROCESSING UNIT
| Base clock | 2235 MHz 745 MHz |
| Boost clock | 2520 MHz 876 MHz |
| Transistors | 76,300 million 7,080 million |
| Process size | 5 nm 28 nm |
| Process type | N/A |
| Die size | 609 mm² 561 mm² |
| MCD Die density | N/A |
| Chip package | BGA-2152 |
| Density | 125.3M / mm² 12.6M / mm² |
GPU MEMORY
| Memory type | GDDR6X GDDR5 |
| Memory size | 24 GB 12 GB |
| Memory bus | 384 bit 384 bit |
| Bandwidth | 1.01 TB/s 288.4 GB/s |
| Memory clock | 1313 MHz 21 Gbps more effective 1502 MHz 6 Gbps more effective |
GRAPHICS FEATURES (API)
| Vulkan | 1.3 1.2.175 |
| DirectX | 12 Ultimate (12_2) 12 (11_0) |
| OpenGL | 4.6 4.6 |
| OpenCL | 3.0 3.0 |
| Display port | N/A N/A |
| Shader model | 6.8 5.1 |
| Ray tracing | N/A N/A |
| CUDA | 8.9 |
CONFIGURATION
| ROPs | 176 48 |
| Shading units | 16384 2880 |
| TMUs | 512 240 |
| Compute units | N/A |
| L1 cache | 128 KB (per SM) 16 KB (per SMX) |
| L2 cache | 72 MB 1536 KB |
GPU PERFORMANCE
| Pixel rate | 443.5 GPixel/s 52.56 GPixel/s |
| Texture rate | 1,290 GTexel/s 210.2 GTexel/s |
| FP16 (half) | 82.58 TFLOPS (1:1) N/A |
| FP32 (float) | 82.58 TFLOPS 5.046 TFLOPS |
| FP64 (double) | 1,290 GFLOPS (1:64) 1.682 TFLOPS (1:3) |
|
