NVIDIA A100 SXM4 80 GB vs NVIDIA GeForce RTX 3070 Ti
Graphics Processor
| GPU model | GA100 | GA104 |
|---|---|---|
| Architecture | Ampere | Ampere |
| Foundry | TSMC | Samsung |
| Manufacturing process | 7 nm | 8 nm |
| Number of transistors | 54,200 million | 17,400 million |
| Die size | 826 mm² | 392 mm² |
| GPU variant | — | GA104-400-A1 |
Graphics Card
| Release date | Nov 16th, 2020 | May 31st, 2021 |
|---|---|---|
| Generation | Tesla | GeForce 30 |
| Production | Active | Active |
| Interface | PCIe 4.0 x16 | PCIe 4.0 x16 |
| gpu.details.availability | — | Jun 10th, 2021 |
| Predecessor | — | GeForce 20 |
| Launch price | — | 599 USD |
| Reviews | — | 43 in our database |
Clocks
| Base clock | 1275 MHz | 1575 MHz |
|---|---|---|
| Boost clock | 1410 MHz | 1770 MHz |
| Memory clock | 1593 MHz 3.2 Gbps effective | 1188 MHz 19 Gbps effective |
Memory Configuration
| Memory size | 80 GB | 8 GB |
|---|---|---|
| Memory type | HBM2e | GDDR6X |
| Memory bus width | 5120 bit | 256 bit |
| Bandwidth | 2,039 GB/s | 608.3 GB/s |
Render Configuration
| Shading units | 6912 | 6144 |
|---|---|---|
| TMUs | 432 | 192 |
| ROPs | 160 | 96 |
| SM count | 108 | 48 |
| Tensor cores | 432 | 192 |
| Cache L1 | 192 KB (per SM) | 128 KB (per SM) |
| Cache L2 | 40 MB | 4 MB |
| RT cores | — | 48 |
Performance
| Pixel rate | 225.6 GPixel/s | 169.9 GPixel/s |
|---|---|---|
| Texture rate | 609.1 GTexel/s | 339.8 GTexel/s |
| FP16 (half) performance | 77.97 TFLOPS (4:1) | 21.75 TFLOPS (1:1) |
| FP32 (float) performance | 19.49 TFLOPS | 21.75 TFLOPS |
| FP64 (double) performance | 9.746 TFLOPS (1:2) | 339.8 GFLOPS (1:64) |
Dimensions & Outputs
| Slot width | IGP | Dual-slot |
|---|---|---|
| TDP | 400 W | 290 W |
| Suggested PSU | 800 W | 600 W |
| Outputs | No outputs | 1x HDMI3x DisplayPort |
| Power connectors | None | 1x 12-pin |
| Length | — | 267 mm 10.5 inches |
| Width | — | 112 mm 4.4 inches |
| Board number | — | PG143 SKU 15 |
API Support & Features
| DirectX | — | 12 Ultimate (12_2) |
|---|---|---|
| OpenGL | — | 4.6 |
| OpenCL | 3.0 | 3.0 |
| Vulkan | — | 1.2 |
| CUDA | 8.0 | 8.6 |
| Shader model | — | 6.6 |
