NVIDIA A2 vs NVIDIA Quadro RTX 4000
Graphics Processor
| GPU model | GA107 | TU104 |
|---|---|---|
| Architecture | Ampere | Turing |
| Foundry | Samsung | TSMC |
| Manufacturing process | 8 nm | 12 nm |
| Number of transistors | unknown | 13,600 million |
| Die size | unknown | 545 mm² |
| GPU variant | — | TU104-850-A1 |
Graphics Card
| Release date | Nov 10th, 2021 | Nov 13th, 2018 |
|---|---|---|
| Generation | Quadro | Quadro |
| Production | Active | Active |
| Interface | PCIe 4.0 x8 | PCIe 3.0 x16 |
| Launch price | — | 899 USD |
Clocks
| Base clock | 1440 MHz | 1005 MHz |
|---|---|---|
| Boost clock | 1770 MHz | 1545 MHz |
| Memory clock | 1563 MHz 12.5 Gbps effective | 1625 MHz 13 Gbps effective |
Memory Configuration
| Memory size | 16 GB | 8 GB |
|---|---|---|
| Memory type | GDDR6 | GDDR6 |
| Memory bus width | 128 bit | 256 bit |
| Bandwidth | 200.1 GB/s | 416.0 GB/s |
Render Configuration
| Shading units | 1280 | 2304 |
|---|---|---|
| TMUs | 40 | 144 |
| ROPs | 32 | 64 |
| SM count | 10 | 36 |
| Tensor cores | 40 | 288 |
| RT cores | 10 | 36 |
| Cache L1 | 128 KB (per SM) | 64 KB (per SM) |
| Cache L2 | 2 MB | 4 MB |
Performance
| Pixel rate | 56.64 GPixel/s | 98.88 GPixel/s |
|---|---|---|
| Texture rate | 70.80 GTexel/s | 222.5 GTexel/s |
| FP16 (half) performance | 4.531 TFLOPS (1:1) | 14.24 TFLOPS (2:1) |
| FP32 (float) performance | 4.531 TFLOPS | 7.119 TFLOPS |
| FP64 (double) performance | 70.80 GFLOPS (1:64) | 222.5 GFLOPS (1:32) |
Dimensions & Outputs
| Slot width | Single-slot | Single-slot |
|---|---|---|
| TDP | 60 W | 160 W |
| Suggested PSU | 250 W | 450 W |
| Outputs | No outputs | 3x DisplayPort1x USB Type-C |
| Power connectors | None | 1x 8-pin |
| Board number | PG179 SKU 220 | — |
| Length | — | 241 mm 9.5 inches |
| Width | — | 111 mm 4.4 inches |
API Support & Features
| DirectX | 12 Ultimate (12_2) | 12 Ultimate (12_2) |
|---|---|---|
| OpenGL | 4.6 | 4.6 |
| OpenCL | 3.0 | 3.0 |
| Vulkan | 1.2 | 1.2 |
| CUDA | 8.6 | 7.5 |
| Shader model | 6.6 | 6.6 |
