GPU model | NV28 | GF106 |
---|---|---|
GPU variant | NV28M A1 | GF106-875-KA-A1 |
Architecture | Kelvin | Fermi |
Foundry | TSMC | TSMC |
Manufacturing process | 150 nm | 40 nm |
Number of transistors | 36 million | 1,170 million |
Die size | 101 mm² | 238 mm² |
Release date | Mar 15th, 2003 | Oct 5th, 2011 |
---|---|---|
Generation | GeForce 4 Ti | Quadro |
Production | End-of-life | End-of-life |
Interface | AGP 8x | PCIe 2.0 x16 |
Predecessor | GeForce 4 MX | — |
Successor | GeForce FX | — |
Launch price | — | 599 USD |
Reviews | — | 26 in our database |
GPU clock | 300 MHz | 625 MHz |
---|---|---|
Memory clock | 325 MHz 650 Mbps effective | 650 MHz 2.6 Gbps effective |
Shader clock | — | 1250 MHz |
Memory size | 128 MB | 1024 MB |
---|---|---|
Memory type | DDR | GDDR5 |
Memory bus width | 128 bit | 128 bit |
Bandwidth | 10.40 GB/s | 41.60 GB/s |
Pixel shaders | 4 | — |
---|---|---|
Vertex shaders | 2 | — |
TMUs | 8 | 32 |
ROPs | 4 | 16 |
Shading units | — | 192 |
SM count | — | 4 |
Cache L1 | — | 64 KB (per SM) |
Cache L2 | — | 256 KB |
Pixel rate | 1.200 GPixel/s | 5.000 GPixel/s |
---|---|---|
Vertex rate | 150.0 MVertices/s | — |
Texture rate | 2.400 GTexel/s | 20.00 GTexel/s |
FP32 (float) performance | — | 480.0 GFLOPS |
FP64 (double) performance | — | 40.00 GFLOPS (1:12) |
Slot width | Single-slot | Single-slot |
---|---|---|
Length | 216 mm 8.5 inches | 178 mm 7 inches |
TDP | unknown | 62 W |
Suggested PSU | 200 W | 250 W |
Outputs | 1x DVI1x VGA1x S-Video | 2x DVI |
Power connectors | None | None |
Width | — | 111 mm 4.4 inches |
Board number | — | P1232 |
DirectX | 8.1 | 12 (11_0) |
---|---|---|
OpenGL | 1.5 | 4.6 |
OpenCL | — | 1.1 |
Vulkan | — | — |
Pixel shader | 1.3 | — |
Vertex shader | 1.1 | — |
CUDA | — | 2.1 |
Shader model | — | 5.1 |