GPU model | Ellesmere | GP104 |
---|---|---|
GPU variant | Ellesmere XT | — |
Architecture | GCN 4.0 | Pascal |
Foundry | GlobalFoundries | TSMC |
Manufacturing process | 14 nm | 16 nm |
Number of transistors | 5,700 million | 7,200 million |
Die size | 232 mm² | 314 mm² |
Release date | Aug 4th, 2016 | Feb 21st, 2018 |
---|---|---|
Generation | Mobility Radeon (RX M400) | Quadro Mobile (Px200) |
Production | Active | End-of-life |
Interface | MXM-B (3.0) | MXM-B (3.0) |
Predecessor | Crystal System | — |
Base clock | 1000 MHz | 1328 MHz |
---|---|---|
Boost clock | 1077 MHz | 1543 MHz |
Memory clock | 2000 MHz 8 Gbps effective | 1753 MHz 7 Gbps effective |
Memory size | 8 GB | 6 GB |
---|---|---|
Memory type | GDDR5 | GDDR5 |
Memory bus width | 256 bit | 192 bit |
Bandwidth | 256.0 GB/s | 168.3 GB/s |
Shading units | 2304 | 1792 |
---|---|---|
TMUs | 144 | 112 |
ROPs | 32 | 64 |
Compute units | 36 | — |
Cache L1 | 16 KB (per CU) | 48 KB (per SM) |
Cache L2 | 2 MB | 1536 KB |
SM count | — | 14 |
Pixel rate | 34.46 GPixel/s | 98.75 GPixel/s |
---|---|---|
Texture rate | 155.1 GTexel/s | 172.8 GTexel/s |
FP16 (half) performance | 4.963 TFLOPS (1:1) | 86.41 GFLOPS (1:64) |
FP32 (float) performance | 4.963 TFLOPS | 5.530 TFLOPS |
FP64 (double) performance | 310.2 GFLOPS (1:16) | 172.8 GFLOPS (1:32) |
Slot width | MXM Module | MXM Module |
---|---|---|
TDP | 100 W | 75 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 | 6.4 |
CUDA | — | 6.1 |