Nvidia GTX 1070 Specs Revealed – 1.6 GHz Boost Clock, 8GB GDDR5 Memory, and 150W TDP
Ever since Nvidia released their new Pascal based GTX 10-Series graphics cards, most of the attention is focused on the flagship GeForce GTX 1080, while there is very little that we know about the younger sibling GTX 1070. But that was until now as the chip maker has revealed specifications of its cheaper solution, as well.
As expected, Nvidia GTX 1070 is based on a cut down 16nm GP104 silicon, called the “GP104-200-A1” ASIC. The GPU features 1,920 CUDA cores (down from 2,560 on the GTX 1080) along with 15 streaming processors, 120 TMUs and 64 ROPs.
The company has also lowered the GPU boost clock speeds, down to 1,600 MHz from 1,733 MHz. Further, the 1070 features 8 GB of GDDR5 memory across a 256-bit bus, delivering a total bandwidth of 256 GB/s.
The card pushes around 6.75 TFLOPs of compute performance from a chip built around 7.2 billion transistors, as well as supports 2-way SLI configurations with SLI-HB bridge support.
TDP sits at 150W which is powered through a single 8-pin PCIe power connector. Display outputs include 3 Display Port 1.4, HDMI 2.0b and a dual-DVI-D port.
Nvidia GTX 1070 launches on June 10th for an MSRP of $379, which is $50 more than the launch-price of GTX 970. Nvidia also plans to launch a “Founders Edition” of the card for a higher cost of $450 this summer.