Nvidia makes Geforce Titan with Kepler GK110

When Nvidia released the Geforce GTX 680 in March 2012, it was clear that the new flagship was not based on the full version of the Kepler graphics processor (GK110). Instead, a slimmed-down variant (GK104) was used to keep production costs down. Now, almost a year later, the full-fledged Kepler may be on its way to a consumer-class Geforce.

Several independent sources tell SweClockers that GK110 is appearing in Geforce Titan – an upcoming graphics card in the worsting class. The name alludes to the world’s fastest supercomputer Cray Titan at the Oak Ridge National Laboratory in the USA, which is based on 18,688 Nvidia Tesla K20X with just GK110.

The Nvidia Tesla K20X computing card is based on the GK110 with 2,688 CUDA cores, 384-bit memory bus and 6 GB of GDDR5 memory. The circuit actually contains 2,880 computing devices, but Nvidia has disabled a cluster (SMX), probably for manufacturing reasons. At the same time, the clock frequencies remain at relatively low 732 MHz GPU and 5.2 GHz GDDR5.

According to SweClocker’s sources, the launch of Geforce Titan will be similar to that of the Geforce GTX 690. Partner manufacturers must follow Nvidia’s reference design to the letter and may not even put their own stickers on the graphics cards. Performance is estimated at about 85 percent of a Geforce GTX 690.

The same sources claim that the Geforce Titan will be released at the end of February and has a recommended price of 899 USD.