While we already know the actual performance of the GeForce GTX 1080, now it comes down to the GeForce GTX 1070. In official presentation NVIDIA”s CEO presented very little information about GeForce GTX 1070, but today we know its full specifications. The NVIDIA GeForce GTX 1070 is based on Pascal GP104-200 silicon manufactured @16nm. It will consist of 1920 CUDA Cores with 120 TMUs that could keep 64 ROPs, all this at a Turbo frequency 1600 MHz with 8 GB of GDDR5 @ 8000 MHz memory linked to a memory interface of 256-bits which will give a bandwidth of 256 GB / s with a TDP of 150W, so it will require one PCI-Express 8 – pin connector for power.

NVIDIA GeForce GTX 1070 Official Specifications: 1920 CUDA Cores and 1.60 GHz Boost Clock


Official specifications of the GeForce GTX 1070 are completed with four video outputs as three DisplayPort 1.4, HDMI 2.0B and a DVI. NVIDIA set the MSRP for the GeForce GTX 1070 at US $379, $50 higher than the launch-price of the GTX 970, the most popular SKU based on the GM204 silicon.  The company also plans to sell a premium reference-design card, dubbed the GeForce GTX 1070 Founders Edition, at $449, at a $70 premium.

With these data on the table, finally we can say that, GTX 1070 will perform slightly higher than the GeForce GTX 980 and it also consumes 15W, so that the trend of improving performance reducing energy consumption remains. The GPU will hits the shelves on June 10, 2016.

NVIDIA GeForce GTX 1070 Official Specifications

  • 16 nm “GP104” silicon, 7.2 billion transistors, “GP104-200-A1” ASIC
  • 1,920 CUDA cores, 15 out of 20 streaming multiprocessors enabled on the GP104 silicon
  • 120 TMUs, 64 ROPs
  • 256-bit GDDR5 memory, 8 GB standard memory amount
  • Max GPU Boost frequency 1600 MHz
  • 6.75 TFLOP/s single-precision floating point performance
  • 150W TDP, single 8-pin PCIe power connector
  • 3x DisplayPort 1.4, 1x HDMI 2.0b
  • 2-way SLI with SLI HB bridge support
NVIDIA GeForce GTX 1080 Vs GeForce GTX 1070
May 15th 2016GeForce GTX 1080GeForce GTX 1070
ArchitecturePascal 16nm FinFETPascal 16nm FinFET
GPU GP104-400 GP104-200
Streaming Multiprocessors2015
CUDA Cores25601920
Memory Type8GB GDDR5X8GB GDDR5
Base Clock1607 MHz(?)
Boost Clock 1733 MHz1600 MHz
Memory Clock1250 MHz2000 MHz
Effective Memory Clock10000 MHz8000 MHz
Memory Bus256-bit256-bit
Memory Bandwidth320 GB/s256 GB/s
Power Connectors1x 8pin1x 8pin
$699 FE
$449 FE

Via: techpowerup