GPU model | Gemini Lake GT1 | GA102 |
---|---|---|
Architecture | Generation 9.5 | Ampere |
Foundry | Intel | Samsung |
Manufacturing process | 14 nm | 8 nm |
Number of transistors | unknown | 28,300 million |
Die size | unknown | 628 mm² |
Release date | Dec 11th, 2017 | — |
---|---|---|
Generation | HD Graphics-T (Goldmont Plus) | — |
Production | Active | — |
Interface | Ring Bus | — |
Reviews | 1 in our database | — |
Base clock | 200 MHz | 1305 MHz |
---|---|---|
Boost clock | 650 MHz | 1740 MHz |
Memory clock | System Shared | 1812 MHz 14.5 Gbps effective |
Memory size | System Shared | 48 GB |
---|---|---|
Memory type | System Shared | GDDR6 |
Memory bus width | System Shared | 384 bit |
Bandwidth | System Dependent | 695.8 GB/s |
Shading units | 96 | 10752 |
---|---|---|
TMUs | 12 | 336 |
ROPs | 2 | 112 |
Execution units | 12 | — |
SM count | — | 84 |
Tensor cores | — | 336 |
RT cores | — | 84 |
Cache L1 | — | 128 KB (per SM) |
Cache L2 | — | 6 MB |
Pixel rate | 1.300 GPixel/s | 194.9 GPixel/s |
---|---|---|
Texture rate | 7.800 GTexel/s | 584.6 GTexel/s |
FP16 (half) performance | 249.6 GFLOPS (2:1) | 37.42 TFLOPS (1:1) |
FP32 (float) performance | 124.8 GFLOPS | 37.42 TFLOPS |
FP64 (double) performance | 31.20 GFLOPS (1:4) | 1,169 GFLOPS (1:32) |
Slot width | IGP | Dual-slot |
---|---|---|
TDP | 5 W | 300 W |
Outputs | No outputs | 3x DisplayPort |
Length | — | 267 mm 10.5 inches |
Width | — | 112 mm 4.4 inches |
Suggested PSU | — | 700 W |
Power connectors | — | 8-pin EPS |
DirectX | 12 (12_1) | 12 Ultimate (12_2) |
---|---|---|
OpenGL | 4.6 | 4.6 |
OpenCL | 3.0 | 3.0 |
Vulkan | 1.2 | 1.2 |
Shader model | 6.4 | 6.6 |
CUDA | — | 8.6 |
Release date | — | Oct 5th, 2020 |
---|---|---|
gpu.details.availability | — | 2021 |
Generation | — | Tesla |
Production | — | Active |
Interface | — | PCIe 4.0 x16 |