GPU model | Raven | TU116 |
---|---|---|
Architecture | GCN 5.0 | Turing |
Foundry | GlobalFoundries | TSMC |
Manufacturing process | 14 nm | 12 nm |
Number of transistors | 4,940 million | 6,600 million |
Die size | 210 mm² | 284 mm² |
GPU variant | — | TU116-400-A1 |
Release date | Apr 19th, 2018 | — |
---|---|---|
Generation | Raven Ridge (Vega) | — |
Production | Active | — |
Interface | IGP | — |
Predecessor | Great Horned Owl | — |
Successor | Picasso | — |
Base clock | 300 MHz | 1500 MHz |
---|---|---|
Boost clock | 1101 MHz | 1770 MHz |
Memory clock | System Shared | 1500 MHz 12 Gbps effective |
Memory size | System Shared | 6 GB |
---|---|---|
Memory type | System Shared | GDDR6 |
Memory bus width | System Shared | 192 bit |
Bandwidth | System Dependent | 288.0 GB/s |
Shading units | 512 | 1536 |
---|---|---|
TMUs | 32 | 96 |
ROPs | 8 | 48 |
Compute units | 8 | — |
SM count | — | 24 |
Cache L1 | — | 64 KB (per SM) |
Cache L2 | — | 1536 KB |
Pixel rate | 8.808 GPixel/s | 84.96 GPixel/s |
---|---|---|
Texture rate | 35.23 GTexel/s | 169.9 GTexel/s |
FP16 (half) performance | 2.255 TFLOPS (2:1) | 10.87 TFLOPS (2:1) |
FP32 (float) performance | 1,127 GFLOPS | 5.437 TFLOPS |
FP64 (double) performance | 70.46 GFLOPS (1:16) | 169.9 GFLOPS (1:32) |
Slot width | IGP | Dual-slot |
---|---|---|
TDP | 35 W | 120 W |
Outputs | No outputs | 1x DVI1x HDMI1x DisplayPort |
Power connectors | None | 1x 8-pin |
Length | — | 229 mm 9 inches |
Width | — | 111 mm 4.4 inches |
Height | — | 35 mm 1.4 inches |
Suggested PSU | — | 300 W |
Board number | — | PG160 SKU 21 PG161 SKU 21 |
DirectX | 12 (12_1) | 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 |
Release date | — | Feb 22nd, 2019 |
---|---|---|
Generation | — | GeForce 16 |
Predecessor | — | GeForce 10 |
Successor | — | GeForce 20 |
Production | — | Active |
Launch price | — | 279 USD |
Current price | — | Amazon / Newegg |
Interface | — | PCIe 3.0 x16 |
Reviews | — | 59 in our database |