Is NVIDIA going to be releasing their next-generation graphics processor next month? SweClockers is reporting that the NVIDIA GeForce GTX 750 Ti will be released in the middle of February and that it will be using a GPU based on the Maxwell architecture. Maxwell is the successor to Kepler, so this would be a pretty big deal for gamers around the world. The site didn’t give any specific details of the video card, but did mention that it would likely be made by TSMC on the 28nm process and then down the road move it down to 20nm for a product refresh. They also believe that the upcoming NVIDIA Geforce GTX 750 Ti will replace the current NVIDIA Geforce GTX 650 Ti Boost, so that could mean this card would be in the $149-$169 price range as that was the original SRP on the GTX 650 Ti Boost when it was released in March 2013.
Other sites are claiming that the GeForce GTX 750 Ti will be launching on February 18th and will be bringing new features to the table like Unified Virtual Memory (UVM). This makes sense as NVIDIA already announced that Maxwell would be using UVM at the 2013 GPU Technology Conference. The goal of UVM is basically to use page faults in the virtual memory systems to detect when a piece of memory is being accessed on the GPU and move the pages to the device, and then move it back to the CPU when the CPU accesses it. This is something that is kind of being done today with Unified Virtual Addressing (UVA), but Maxwell does it with hardware.
NVIDIA of course hasn’t confirmed any of this and does not talk about leaks or any unreleased products. It does seem odd to us that NVIDIA would place the first Maxwell GPU into the GeForce GTX 700 series, but NVIDIA knows what they are doing. Maybe they are saving the GeForce GTX 800 series for when the shrink the process node.