There have been two camps of rumours on Fermi's launch date. The first have been saying Fermi is all set to launch in November/December, 2009. The second consider it entirely unlikely, and suggest a late Q1 2010 release, at best.

Now, we have the word straight from the boss. Nvidia CEO Jen Hsun-Huang suggests Fermi production will only be ramped in 2010.

More details next page.

Nvidia is in the news again, for all the wrong reasons. Lucid Logix' HYDRA has been a promising and anticipated product ever since it was announced back in 2008. Lucid claimed it would allow graphics cards from different manufacturers to scale better than current AFR solutions such as SLI or Crossfire. So, you could combine a Radeon card with a Geforce (or more cards) and they would scale accordingly, by using methods such as Split Frame Rendering.

Both AMD and Nvidia were understandably skeptical about this solution. Now, Lucid's Hydra 200 chip is ready and available in MSI's Big Bang Fuzion motherboard.

More next page.

Nvidia have been at a war of words with Intel for a while now. Fudzilla have uncovered a website called "Intel's Insides" which features a collection of somewhat humourous comics parodying Intel.

Right below the comics, there's a small Nvidia logo, which has can lead people to believe that it is indeed Nvidia, the visual computing company, getting into a war of images with Intel.

Tick-tock, tick-tock, says the clock. A clock called "Larrabee" attached to what looks like a bomb. So, are images like these really Nvidia's work?


More next page.

This is a rumour that has surfaced and died several times, and it has resurfaced again. However, this time, there is more evidence than usual. Over the last year or two, Nvidia have spent a lot of time and effort into marketing stream computing. GPUs, or parallel processors, are significantly more powerful than their CPU counterparts. However, parallel processors can only process specific instructions, and a CPU is still required for general processing. So, as much as Nvidia would convince consumers about the benefits of GPU, there is no doubt a GPU would be useless were it not run together with a CPU.

Recently, Nvidia reportedly presented a CPU+GPU for 2017. AMD Fusion is all about the same central idea - a GPU and a CPU on one die. Though Intel will be the first with a GPU and a CPU on one package, they will be on different dies, and the GPU will be a weak Intel IGP anyway. Now, rumours are suggesting Nvidia are keen to not be left behind by the CPU+GPU revolution, and are developing x86 CPUs.

More details next page.

While Nvidia and a few other reporters (most notably, Fudzilla) claim Fermi based products are well on their way, a host of other reporters are now claiming Nvidia is "doing their best" to have a working Fermi demonstration at CES 2010, which is scheduled for 7th to 10th November.

Nvidia's original release date was apparently set for late November, but recently Fudzilla reported that this date was shifted to December. Meanwhile, most other sources claimed retail availability of Fermi would only happen in Q1 2010.

More details next page.

Remember the mysterious "new" dual-GPU from EVGA/Nvidia? It is here, and there are no surprises. It is a combination of GTX 275 as the "GPU" and a GTS 250 as the PhysX Processor, "PPU". 

A forum member on Chinese website PCZilla has leaked some stunning images purpotedly rendered on a Fermi GPU. 

AMD's senior manager of development relations, Richard Huddy, in a talk with Hexus, feels Nvidia is "somewhat abandoning the gaming market". 

TSMC's troubled 40nm process has once again hit yield issues. When RV740 (HD 4770) first released in Q2 2009, availability remained poor due to troubled yields reported to be as low as 20-30%. By July, the 40nm yield had improved to a respectable 60%. 

Eight years may be an age (or two) in the computer hardware industry, but Nvidia is nevertheless presenting a product that will be available in 2017.