GPU model | Picasso | GA104 |
---|---|---|
Architecture | GCN 5.0 | Ampere |
Foundry | GlobalFoundries | Samsung |
Manufacturing process | 14 nm | 8 nm |
Number of transistors | 4,940 million | 17,400 million |
Die size | 210 mm² | 392 mm² |
Release date | Jan 6th, 2019 | — |
---|---|---|
Generation | Picasso (Vega Mobile) | — |
Production | Active | — |
Interface | IGP | — |
Predecessor | Raven Ridge | — |
Successor | Renoir | — |
Base clock | 300 MHz | 1020 MHz |
---|---|---|
Boost clock | 1200 MHz | 1545 MHz |
Memory clock | System Shared | 1500 MHz 12 Gbps effective |
Memory size | System Shared | 12 GB |
---|---|---|
Memory type | System Shared | GDDR6 |
Memory bus width | System Shared | 256 bit |
Bandwidth | System Dependent | 384.0 GB/s |
Shading units | 384 | 5632 |
---|---|---|
TMUs | 24 | 176 |
ROPs | 8 | 80 |
Compute units | 6 | — |
SM count | — | 44 |
Tensor cores | — | 176 |
RT cores | — | 44 |
Cache L1 | — | 128 KB (per SM) |
Cache L2 | — | 4 MB |
Pixel rate | 9.600 GPixel/s | 123.6 GPixel/s |
---|---|---|
Texture rate | 28.80 GTexel/s | 271.9 GTexel/s |
FP16 (half) performance | 1.843 TFLOPS (2:1) | 17.40 TFLOPS (1:1) |
FP32 (float) performance | 921.6 GFLOPS | 17.40 TFLOPS |
FP64 (double) performance | 57.60 GFLOPS (1:16) | 543.8 GFLOPS (1:32) |
Slot width | IGP | — |
---|---|---|
TDP | 15 W | 140 W |
Outputs | No outputs | No outputs |
Power connectors | — | None |
DirectX | 12 (12_1) | 12 Ultimate (12_2) |
---|---|---|
OpenGL | 4.6 | 4.6 |
OpenCL | 2.1 | 3.0 |
Vulkan | 1.2 | 1.2 |
Shader model | 6.4 | 6.6 |
CUDA | — | 8.6 |
Release date | — | Unknown |
---|---|---|
Generation | — | Quadro Mobile (Ax000) |
Production | — | Unreleased |
Interface | — | PCIe 4.0 x16 |