GPU model | NV28 | GP10B |
---|---|---|
GPU variant | NV28M A1 | Tegra X2 |
Architecture | Kelvin | Pascal |
Foundry | TSMC | TSMC |
Manufacturing process | 150 nm | 16 nm |
Number of transistors | 36 million | unknown |
Die size | 101 mm² | unknown |
Release date | Mar 15th, 2003 | — |
---|---|---|
Generation | GeForce 4 Ti | — |
Production | End-of-life | — |
Interface | AGP 8x | — |
Predecessor | GeForce 4 MX | — |
Successor | GeForce FX | — |
GPU clock | 300 MHz | — |
---|---|---|
Memory clock | 325 MHz 650 Mbps effective | System Shared |
Base clock | — | 854 MHz |
Boost clock | — | 1465 MHz |
Memory size | 128 MB | System Shared |
---|---|---|
Memory type | DDR | System Shared |
Memory bus width | 128 bit | System Shared |
Bandwidth | 10.40 GB/s | System Dependent |
Pixel shaders | 4 | — |
---|---|---|
Vertex shaders | 2 | — |
TMUs | 8 | 16 |
ROPs | 4 | 16 |
Shading units | — | 256 |
SM count | — | 2 |
Cache L1 | — | 48 KB (per SM) |
Cache L2 | — | 512 KB |
Pixel rate | 1.200 GPixel/s | 23.44 GPixel/s |
---|---|---|
Vertex rate | 150.0 MVertices/s | — |
Texture rate | 2.400 GTexel/s | 23.44 GTexel/s |
FP16 (half) performance | — | 1.500 TFLOPS (2:1) |
FP32 (float) performance | — | 750.1 GFLOPS |
FP64 (double) performance | — | 23.44 GFLOPS (1:32) |
Slot width | Single-slot | IGP |
---|---|---|
Length | 216 mm 8.5 inches | 87 mm 3.4 inches |
TDP | unknown | 15 W |
Suggested PSU | 200 W | — |
Outputs | 1x DVI1x VGA1x S-Video | No outputs |
Power connectors | None | — |
Width | — | 50 mm 2 inches |
DirectX | 8.1 | 12 (12_1) |
---|---|---|
OpenGL | 1.5 | 4.6 |
OpenCL | — | 1.2 |
Vulkan | — | 1.2 |
Pixel shader | 1.3 | — |
Vertex shader | 1.1 | — |
CUDA | — | 6.2 |
Shader model | — | 6.4 |
Release date | — | 2016 |
---|---|---|
Generation | — | Tegra |
Production | — | Active |
Launch price | — | 249 USD |
Interface | — | IGP |