Geforce gtx 1070 more powerful than nvidia titan x

The Nvidia GeForce GTX 1070 promises to come at a much tighter price than the GTX 1080 while still maintaining great performance thanks to its Pascal architecture. The first benchmark confirms that the GeForce GTX 1070 is more powerful than the TITAN X.

GeForce GTX 1070 shows superior to GTX TITAN X in 3D Mark FireStrike

The GeForce GTX 1070 has been passed through 3DMark FireStrike at 1080p, 2K and 4K resolutions showing excellent performance as it is 3% more powerful than the previous top of the range of the North American company, the GeForce GTX TITAN X.

At 1080p resolution (FireStrike) the GTX 1070 has obtained 17557 points compared to 17396 points on TITAN X, we have gone to 2K resolution (FireStrike Advanced) and the GTX 1070 has 8327 points compared to 7989 points on TITAN X, finally at At high 4K resolution (FireStrike Ultra) the GTX 1070 scores 4, 078 points compared to 3, 862 points on the TITAN X.

GeForce GTX 1070 technical characteristics

The GeForce GTX 1070 uses a trimmed variant of the Pascal GP104 GPU with a total of 1, 920 CUDA cores, 120 TMUs, and supposedly the same 64 ROPs as its older sister, although the latter is unconfirmed. This GPU will operate at a maximum frequency of 1.6 GHz and offers a theoretical maximum power of 6.75 TFLOPs. The GPU is accompanied by 8 GB of GDDR5 memory with a 256- bit interface and a bandwidth of 256 GB / s. All this with a reduced TDP of 150W, so Pascal once again boasts a formidable energy efficiency.

We recommend reading our guide to the best graphics cards on the market by ranges

The GeForce GTX 1070 is powered only by an 8-pin connector and offers video outputs in the form of 3x DisplayPort 1.4, HDMI 2.0B, and Dual-DVI to handle up to 5 displays.

The GeForce GTX 1070 will be officially announced on June 10 and its price will be approximately 450 euros.

NVIDIA GeForce GTX 1080 Specifications
	GeForce GTX 1080	GeForce GTX 1070
Architecture	Pascal 16nm FinFET	Pascal 16nm FinFET
GPU	GP104-400	GP104-200
Streaming Multiprocessors	twenty	fifteen
CUDA Cores	2560	1920
TMUs	160	120
ROPs	64	64 (?)
TFLOPs	8.2 TFLOPs	6.5 TFLOPs
Memory type	8GB GDDR5X	8GB GDDR5
Base Clock	1607 MHz	(?)
Boost Clock	1733 MHz	1600 MHz
Memory clock	1250 MHz	2000 MHz
Effective Memory Clock	10000 MHz	8000 MHz
Memory bus	256-bit	256-bit
Memory Bandwidth	320 GB / s	256 GB / s
TDP	180W	150W
Power Connectors	1x 8pin	1x 8pin

Source: techpowerup