GPU model | G92B | GM108 |
---|---|---|
GPU variant | G92-421-B1 | N16V-GMR1-A1 |
Architecture | Tesla | Maxwell |
Foundry | TSMC | TSMC |
Manufacturing process | 55 nm | 28 nm |
Number of transistors | 754 million | unknown |
Die size | 260 mm² | unknown |
Release date | Feb 16th, 2010 | — |
---|---|---|
Generation | GeForce 200 | — |
Production | End-of-life | — |
Interface | PCIe 2.0 x16 | — |
Reviews | 61 in our database | — |
Predecessor | GeForce 9 | — |
Successor | GeForce 400 | — |
GPU clock | 633 MHz | — |
---|---|---|
Shader clock | 1296 MHz | — |
Memory clock | 1134 MHz 2.3 Gbps effective | 1253 MHz 5 Gbps effective |
Base clock | — | 978 MHz |
Boost clock | — | 1006 MHz |
Memory size | 896 MB | 2 GB |
---|---|---|
Memory type | GDDR3 | GDDR5 |
Memory bus width | 448 bit | 64 bit |
Bandwidth | 127.0 GB/s | 40.10 GB/s |
Shading units | 240 | 384 |
---|---|---|
TMUs | 80 | 24 |
ROPs | 28 | 8 |
SM count | 30 | — |
Cache L2 | 224 KB | 1024 KB |
SMM count | — | 3 |
Cache L1 | — | 64 KB (per SMM) |
Pixel rate | 17.72 GPixel/s | 8.048 GPixel/s |
---|---|---|
Texture rate | 50.64 GTexel/s | 24.14 GTexel/s |
FP32 (float) performance | 622.1 GFLOPS | 772.6 GFLOPS |
FP64 (double) performance | 77.76 GFLOPS (1:8) | 24.14 GFLOPS (1:32) |
Slot width | Dual-slot | IGP |
---|---|---|
Length | 267 mm 10.5 inches | — |
TDP | 219 W | 30 W |
Suggested PSU | 550 W | — |
Outputs | 2x DVI | No outputs |
Power connectors | 1x 6-pin + 1x 8-pin | None |
DirectX | 11.1 (10_0) | 12 (11_0) |
---|---|---|
OpenGL | 3.3 | 4.6 |
OpenCL | 1.1 | 3.0 |
Vulkan | — | 1.1 |
CUDA | 1.3 | 5.0 |
Shader model | 4.0 | 5.1 |
Release date | — | Nov 17th, 2017 |
---|---|---|
Generation | — | GeForce MX (1xx) |
Production | — | End-of-life |
Interface | — | PCIe 3.0 x4 |