GPU model | Thames | GM108 |
---|---|---|
GPU variant | Thames XTX | — |
Architecture | TeraScale 2 | Maxwell |
Foundry | TSMC | TSMC |
Manufacturing process | 40 nm | 28 nm |
Number of transistors | 716 million | unknown |
Die size | 104 mm² | unknown |
Release date | Jan 7th, 2013 | Mar 12th, 2014 |
---|---|---|
Generation | London (HD 7600M) | GeForce 800M |
Production | End-of-life | End-of-life |
Interface | PCIe 2.0 x16 | PCIe 3.0 x8 |
Predecessor | Vancouver | GeForce 700M |
Successor | Solar System | GeForce 900M |
GPU clock | 725 MHz | — |
---|---|---|
Memory clock | 900 MHz 3.6 Gbps effective | 900 MHz 1800 Mbps effective |
Base clock | — | 1082 MHz |
Boost clock | — | 1150 MHz |
Memory size | 2 GB | 2 GB |
---|---|---|
Memory type | GDDR5 | DDR3 |
Memory bus width | 128 bit | 64 bit |
Bandwidth | 57.60 GB/s | 14.40 GB/s |
Shading units | 480 | 256 |
---|---|---|
TMUs | 24 | 16 |
ROPs | 8 | 8 |
Compute units | 6 | — |
Cache L1 | 8 KB (per CU) | 64 KB (per SMM) |
Cache L2 | 256 KB | 1024 KB |
SMM count | — | 2 |
Pixel rate | 5.800 GPixel/s | 9.200 GPixel/s |
---|---|---|
Texture rate | 17.40 GTexel/s | 18.40 GTexel/s |
FP32 (float) performance | 696.0 GFLOPS | 588.8 GFLOPS |
FP64 (double) performance | — | 18.40 GFLOPS (1:32) |
TDP | 25 W | 33 W |
---|---|---|
Outputs | No outputs | No outputs |
DirectX | 11.2 (11_0) | 12 (11_0) |
---|---|---|
OpenGL | 4.4 | 4.6 |
OpenCL | 1.2 | 3.0 |
Vulkan | — | 1.1 |
Shader model | 5.0 | 5.1 |
CUDA | — | 5.0 |