GPU model | Scrapper | GF108 |
---|---|---|
Architecture | TeraScale 3 | Fermi |
Foundry | TSMC | TSMC |
Manufacturing process | 32 nm | 40 nm |
Number of transistors | 1,303 million | 585 million |
Die size | 246 mm² | 116 mm² |
GPU variant | — | N11P-GE-A1 |
Release date | Dec 28th, 2013 | — |
---|---|---|
Generation | Richland (HD 8000) | — |
Production | End-of-life | — |
Interface | IGP | — |
Predecessor | Trinity | — |
Successor | Kabini | — |
Base clock | 720 MHz | — |
---|---|---|
Boost clock | 844 MHz | — |
Memory clock | System Shared | 800 MHz 1600 Mbps effective |
GPU clock | — | 500 MHz |
Shader clock | — | 1000 MHz |
Memory size | System Shared | 512 MB |
---|---|---|
Memory type | System Shared | DDR3 |
Memory bus width | System Shared | 128 bit |
Bandwidth | System Dependent | 25.60 GB/s |
Shading units | 384 | 96 |
---|---|---|
TMUs | 24 | 16 |
ROPs | 8 | 4 |
Compute units | 4 | — |
Cache L1 | 16 KB (per CU) | 64 KB (per SM) |
Cache L2 | 256 KB | 256 KB |
SM count | — | 2 |
Pixel rate | 6.752 GPixel/s | 2.000 GPixel/s |
---|---|---|
Texture rate | 20.26 GTexel/s | 8.000 GTexel/s |
FP32 (float) performance | 648.2 GFLOPS | 192.0 GFLOPS |
FP64 (double) performance | — | 16.00 GFLOPS (1:12) |
Slot width | IGP | — |
---|---|---|
TDP | 65 W | 23 W |
Outputs | No outputs | No outputs |
DirectX | 11.2 (11_0) | 12 (11_0) |
---|---|---|
OpenGL | 4.4 | 4.6 |
OpenCL | 1.2 | 1.1 |
Vulkan | — | — |
Shader model | 5.0 | 5.1 |
CUDA | — | 2.1 |
Release date | — | Sep 3rd, 2010 |
---|---|---|
Generation | — | GeForce 400M |
Production | — | End-of-life |
Interface | — | PCIe 2.0 x16 |
Predecessor | — | GeForce 300M |
Successor | — | GeForce 500M |