NVIDIA Tesla P100 PCIe 16 GB vs NVIDIA Tesla T4
Graphics Processor
| GPU model | GP100 | TU104 |
|---|---|---|
| GPU variant | GP100-893-A1 | TU104-895-A1 |
| Architecture | Pascal | Turing |
| Foundry | TSMC | TSMC |
| Manufacturing process | 16 nm | 12 nm |
| Number of transistors | 15,300 million | 13,600 million |
| Die size | 610 mm² | 545 mm² |
Graphics Card
| Release date | Jun 20th, 2016 | Sep 13th, 2018 |
|---|---|---|
| Generation | Tesla | Tesla |
| Production | End-of-life | Active |
| Launch price | 5,699 USD | — |
| Interface | PCIe 3.0 x16 | PCIe 3.0 x16 |
Clocks
| Base clock | 1190 MHz | 585 MHz |
|---|---|---|
| Boost clock | 1329 MHz | 1590 MHz |
| Memory clock | 715 MHz 1430 Mbps effective | 1250 MHz 10 Gbps effective |
Memory Configuration
| Memory size | 16 GB | 16 GB |
|---|---|---|
| Memory type | HBM2 | GDDR6 |
| Memory bus width | 4096 bit | 256 bit |
| Bandwidth | 732.2 GB/s | 320.0 GB/s |
Render Configuration
| Shading units | 3584 | 2560 |
|---|---|---|
| TMUs | 224 | 160 |
| ROPs | 96 | 64 |
| SM count | 56 | 40 |
| Cache L1 | 24 KB (per SM) | 64 KB (per SM) |
| Cache L2 | 4 MB | 4 MB |
| Tensor cores | — | 320 |
| RT cores | — | 40 |
Performance
| Pixel rate | 127.6 GPixel/s | 101.8 GPixel/s |
|---|---|---|
| Texture rate | 297.7 GTexel/s | 254.4 GTexel/s |
| FP16 (half) performance | 19.05 TFLOPS (2:1) | 65.13 TFLOPS (8:1) |
| FP32 (float) performance | 9.526 TFLOPS | 8.141 TFLOPS |
| FP64 (double) performance | 4.763 TFLOPS (1:2) | 254.4 GFLOPS (1:32) |
Dimensions & Outputs
| Slot width | Dual-slot | Single-slot |
|---|---|---|
| Length | 267 mm 10.5 inches | 168 mm 6.6 inches |
| TDP | 250 W | 70 W |
| Suggested PSU | 600 W | 250 W |
| Outputs | No outputs | No outputs |
| Power connectors | 1x 8-pin | None |
| Board number | — | PG183 SKU 200 |
API Support & Features
| DirectX | 12 (12_1) | 12 Ultimate (12_2) |
|---|---|---|
| OpenGL | 4.6 | 4.6 |
| OpenCL | 3.0 | 3.0 |
| Vulkan | 1.2 | 1.2 |
| CUDA | 6.0 | 7.5 |
| Shader model | 6.4 | 6.6 |
