GPU model | Ellesmere | TU116 |
---|---|---|
GPU variant | Ellesmere XT | N18P-G61 |
Architecture | GCN 4.0 | Turing |
Foundry | GlobalFoundries | TSMC |
Manufacturing process | 14 nm | 12 nm |
Number of transistors | 5,700 million | 6,600 million |
Die size | 232 mm² | 284 mm² |
Release date | Mar 1st, 2017 | Apr 23rd, 2020 |
---|---|---|
Generation | Radeon Pro Mobile (WX x100) | GeForce 16 Mobile |
Production | Active | Active |
Interface | PCIe 3.0 x16 | PCIe 3.0 x16 |
Predecessor | — | GeForce 10 Mobile |
Successor | — | GeForce 20 Mobile |
Base clock | 1188 MHz | 1350 MHz |
---|---|---|
Boost clock | 1243 MHz | 1485 MHz |
Memory clock | 1250 MHz 5 Gbps effective | 1500 MHz 12 Gbps effective |
Memory size | 8 GB | 4 GB |
---|---|---|
Memory type | GDDR5 | GDDR6 |
Memory bus width | 256 bit | 128 bit |
Bandwidth | 160.0 GB/s | 192.0 GB/s |
Shading units | 2304 | 1024 |
---|---|---|
TMUs | 144 | 64 |
ROPs | 32 | 32 |
Compute units | 36 | — |
Cache L1 | 16 KB (per CU) | 64 KB (per SM) |
Cache L2 | 2 MB | 1024 KB |
SM count | — | 16 |
Pixel rate | 39.78 GPixel/s | 47.52 GPixel/s |
---|---|---|
Texture rate | 179.0 GTexel/s | 95.04 GTexel/s |
FP32 (float) performance | 5.728 TFLOPS | 3.041 TFLOPS |
FP64 (double) performance | 358.0 GFLOPS (1:16) | 95.04 GFLOPS (1:32) |
FP16 (half) performance | — | 6.083 TFLOPS (2:1) |
Slot width | MXM Module | — |
---|---|---|
Length | 25 mm 1 inches | — |
TDP | 130 W | 50 W |
Outputs | No outputs | No outputs |
Power connectors | None | None |
DirectX | 12 (12_0) | 12 (12_1) |
---|---|---|
OpenGL | 4.6 | 4.6 |
OpenCL | 2.1 | 3.0 |
Vulkan | 1.2 | 1.2 |
Shader model | 6.4 | 6.6 |
CUDA | — | 7.5 |