NVIDIA HAS UNVEILED what it dubs its most powerful graphics card ever, the Geforce GTX Titan Z.
On stage at the GPU Technology Conference (GTC) this week, Nvidia co-founder and CEO Jen Hsun Huang said the graphics card is built around two Kepler GPUs and 12GB of dedicated frame buffer memory, engineered for next-generation 4K displays and multi-monitor gaming.
"If you're in desperate need of a super computer, and you need one really close by and handy and you want it to sit next to your desk, we have just the card for you, Titan Z," said Huang.
"It takes every single video game you have, and allows you to drive it at the highest resolution, with all the features turned on, at 4K resolution on multiple displays and enjoy it at the maximum frame rate."
Expected to arrive sometime in April, the Geforce GTX Titan Z GPU has 5,760 cores, 12GB of memory and eight TFLOPS of performance. Unlike traditional dual-GPU cards, its twin GPUs are tuned to run at the same clock speed with dynamic power balancing, so neither GPU creates a performance bottleneck.
Huang said that researchers, designers and artists will use the Geforce GTX Titan Z for CUDA research, machine learning and developing software and video editing, because "it is a supercomputer you can buy on retail".
"You can buy this supercomputer literally from any country, from any computer OEM in the world," Huang added.
The reason Nvidia built the Geforce GTX Titan Z is because the market wanted so much more performance, Huang claimed.
"If you put it in perspective, [it's] all for $3,000, about 50 cents per processor," Huang explained. "If you compare it to the server that ran the Google Brain, this processor can now fit the entire Google Brain servers in just three GPUs: 17,280 cores, [and at] just 200 Watts, [costing] $12,000 and several hundreds times more energy efficient."
Huang also showed off some gameplay powered by its next generation Geforce GTX Titan Z graphics card running the Unreal Engine 4 gaming engine, which we have to admit looked jaw droopingly realistic. But at $3,000, or about £1,815, those graphics won't come cheap.
Check out the launch video below.
Earlier Tuesday evening, Huang also unveiled Nvidia is working on 3D memory for its next generation Pascal graphics processing units (GPUs), which are set to arrive in "just a couple of years", to solve bandwidth bottleneck problems.
Dubbed Pascal, Nvidia's next-generation GPU technology is named after Blaise Pascal (1623-1662), a French mathematician and physicist, and the genius behind Pascal's Theorem, Pascal's Law and probability theory. µ
Sign up for INQbot – a weekly roundup of the best from the INQ