That is totally irrelevant. Nvidia had delays on GF100 because it is an almost entirely new architecture incorporating some very advanced circuitry such as out of order execution and the PolyMorph engine, among other things. To top things off, it was created on a very bad process node from TSMC.
Next time around, they will be dealing with a die shrink of the same architecture which they have hopefully worked out a lot of bugs and manufacturing kinks at this point. Provided TSMC's 28nm process isn't an epic failure like 40nm, then it is very reasonable to expect a Fermi refresh before the end of the year.
EDIT:
Also the only "official" announcment on the launch of Fermi based GeForce cards by Nvidia was Q1 10 which was just changed to Q2 10 which makes April a strong possibility. All the other "announcements" were simply unofficial leaks.






Bookmarks