Rumors are that NVIDIA plans on releasing the GeForce GTX 750 Ti in February, but what is more interesting is that it is said to be based in Maxwell GPU architecture, rather than previously reported Kepler architecture.
It is really interesting that NVIDIA is using the 700 series for the new Maxwell lineup, instead of the 800 series. The mobility series however is using the 800 series brand name. However, according to WCCF, NVIDIA might be using the 700 series lineup as an entry-level for the Maxwell cards.
Currently, the 700 series consist of the GeForece 780 Ti, GeForece 780, GeForece 770, GeForece 760. A GeForce 750 Ti would make sense for an early low-end Maxwell launch. It would be in range of the 700 series performance, while demoing what users should expect from the later more high-end models that will come out under the 800 series lineup.
If the rumors prove to be true and the GeForce 750 Ti will be based on Maxwell chipset, it will be the first GPU architecture to feature Unified Virtual Memory which allows the GPU and CPU to share the address space. The hardware level integration helps improves memory management and reduces overhead. In addition to that, Maxwell based GPUs would be the first to integrate the Denver CPU. The custom 64-bit ARM Dual-Core CPU will be fused on the PCB and would enhance the GPGPU workload by shifting load from the CPU on to the custom ARM cores. It will also be useful when running NVIDIA’s next generation FLEX (Unified GPU PhysX) processes.
More reports state that the Maxwell desktop chip codename will be GM107 / GM117 and would replace GK106 in terms of performance. The GeForce GTX 750 Ti is rumored to launch in late February and would replace the GeForce GTX 650 Ti Boost for a similar price range.