Fermi is Nvidia's next big thing. Nvidia bills Fermi as an entirely new ground-up design that will finally realize the potential of GPU computing – using the graphics card for more than just pushing pixels and driving them to your display.
"It is completely clear that GPUs are now general purpose parallel computing processors with amazing graphics, and not just graphics chips anymore," said Jen-Hsun Huang, co-founder and CEO of Nvidia. "The Fermi architecture, the integrated tools, libraries and engines are the direct results of the insights we have gained from working with thousands of CUDA developers around the world. We will look back in the coming years and see that Fermi started the new GPU industry."
The Fermi GPU will pack 3 billion transistors, making it one mammoth chip - bigger than anything from ATI - but with that ambition comes challenges.
If this Digitimes report is to be believed, Nvidia has hit a snag or two in its plans to launch Fermi and has now pushed back the date to March 2010.
Digitimes' sources also noted that ATI is expected to launch the 40nm Radeon HD 5670/5570 (Redwood) and HD 5450 (Cedar) GPUs at the end of January or in February 2010.
Update: Nvidia responded to this report saying that it announced in a previous conference call that Fermi would ship sometime in the first quarter of 2010, meaning that there hasn't been any real delay in release time frame.
“We are excited about the attention surrounding GF100 as it forms the foundation for the fastest and most feature rich graphics cards ever designed," said Nvidia's Bryan Del Rizzo. "Great things come to those that wait, and the wait will be over in Q1 as we have said previously. The release date has not changed since our initial comments last November."