GPU model | GF108 | GP108 |
---|---|---|
Architecture | Fermi | Pascal |
Foundry | TSMC | Samsung |
Manufacturing process | 40 nm | 14 nm |
Number of transistors | 585 million | 1,800 million |
Die size | 116 mm² | 74 mm² |
GPU variant | — | GP108-655-A1 (N17S-G3-A1) |
Release date | May 4th, 2012 | Feb 10th, 2020 |
---|---|---|
Generation | GeForce 600M | GeForce MX (3xx) |
Production | End-of-life | End-of-life |
Interface | PCIe 2.0 x16 | PCIe 3.0 x4 |
Reviews | 2 in our database | — |
Predecessor | GeForce 500M | — |
Successor | GeForce 700M | — |
GPU clock | 753 MHz | — |
---|---|---|
Shader clock | 1505 MHz | — |
Memory clock | 785 MHz 3.1 Gbps effective | 1752 MHz 7 Gbps effective |
Base clock | — | 1531 MHz |
Boost clock | — | 1594 MHz |
Memory size | 2 GB | 2 GB |
---|---|---|
Memory type | GDDR5 | GDDR5 |
Memory bus width | 128 bit | 64 bit |
Bandwidth | 50.24 GB/s | 56.06 GB/s |
Shading units | 96 | 384 |
---|---|---|
TMUs | 16 | 24 |
ROPs | 4 | 16 |
SM count | 2 | 3 |
Cache L1 | 64 KB (per SM) | 48 KB (per SM) |
Cache L2 | 256 KB | 512 KB |
Pixel rate | 3.012 GPixel/s | 25.50 GPixel/s |
---|---|---|
Texture rate | 12.05 GTexel/s | 38.26 GTexel/s |
FP32 (float) performance | 289.0 GFLOPS | 1,224 GFLOPS |
FP64 (double) performance | 24.08 GFLOPS (1:12) | 38.26 GFLOPS (1:32) |
FP16 (half) performance | — | 19.13 GFLOPS (1:64) |
TDP | 32 W | 10 W |
---|---|---|
Outputs | No outputs | No outputs |
Slot width | — | IGP |
Power connectors | — | None |
DirectX | 12 (11_0) | 12 (12_1) |
---|---|---|
OpenGL | 4.6 | 4.6 |
OpenCL | 1.1 | 3.0 |
Vulkan | — | 1.2 |
CUDA | 2.1 | 6.1 |
Shader model | 5.1 | 6.4 |