IMO, at this point people who say these things either A) do know the point of the cards and just want to say they don't because they are jealous or B) seriously don't know the point and are behind the times on current graphics options.
I have been saying for over a year now. These cards were not made to be marketed to you if you do not know what you would do with the graphics power. They are marketed towards larger resolutions - which will see a great benefit from these cards. I assume your resolution is not larger that 1980x1200 in which case, the card is not marketed towards you. If your resolution was higher you would understand the NEED for this type of horsepower.
This.
I am currently running 560 Ti SLI (2GB models), because in some games (basically racing sims and BF3), I run a tri-monitor 5760x1080 resolution (or 6060x1080 BC'd).
I really, really hope to see BigK released soon. Based on the current architecture, it's basically guaranteed BigK will be able to smoke every video card out there, which when you are looking for performance, is a nice thing.

(this isn't about fanboyism)
When looking at what the 680 is capable of, and then looking at the architecture and realizing it is far smaller and less performance-packed than what the massive GPU version will likely feature, it shouldn't be hard at all to imagine the performance capability of their top-end massive chip.
A 7 Billion Transistor Kepler?
If I can get a card that is A) a single-GPU card, that can run triple-monitor, and B) absolutely smashes my current setup and gives me even more eye candy for games at that resolution, I'd really, really want it.
I'm fairly confident BigK will allow me to replace a SLI setup with a single-GPU, and increase visual details at the same time. That is a Win/Win all around - SLI is great most of the time, but micro-stutter pops up from time to time between different games and driver versions, and removing that issue would be awesome.
edit:
If this pans out, it could be an interesting switch to Nvidia's previous product strategy. The x80 cards used to be the top-end single-GPU card featuring their largest GPU (by transistor count/die size), and the x90 cards were dual-GPU typically featuring slightly smaller and tweaked GPUs.
But with Nvidia readily admitting they were underwhelmed by AMD's competition this generation, they are capitalizing on it as best they can. They made their 680 using a smaller GPU that likely would have been their 660 or 670, because that ensured their 680 was around the same performance capability of the competition's top card. And they could price it at a point that allowed probably record profit-margin since it wasn't their most biggest chip around the $500 price tag. BigK, if released as the 680, probably would have been $600, and bring in less of a profit-margin just so they could have a competitive price point (but still, probably a healthy margin any way you look at it).
Now, since they are simply pricing and adjusting strategy according to the competition, instead of keeping everything the same and blowing them out of the water (with less profits), they've reserved their largest GPU for the "go big or go home" video card. They'll be able to pull off a beastly 690 if they use BigK for that product, and have the same performance differentials between the x80/x90 products yet do so with a single GPU this time around.
Hopefully they don't price it at $800 like recent x90 cards, although it would be justified for sure based on the current market, and allow them the same very healthy margins they are likely seeing with the GK104 in the 680. They could stand to gain far more market-share if they price it a little more competitively, and still see better margins than if they stuck to their original product strategy. A $650 GTX 690 would be HUGE, especially if the performance matches up to expectations for that chip. Depending I how much I can sell my 560 Ti 2GB cards for, I might even settle for $700 out the gates.
Otherwise I'll probably wait until AMD launches a new generation, and settle for a [hopefully] reduced entry cost for the 690.
Of course, they could do something totally opposite to what I'm predicting, but so far this strategy just seems far too fitting.