Survey time: How many people actually own and regularly use a monitor with a resolution of 1920 x 1200 or larger?
How many people actually play games at such high resolutions?
I'm willing to bet that only a very select few actually do own and regularly use monitors with such gigantic resolutions. While we probably all would want to have a monitor that could support high resolutions, much more be able to play our beloved video games on them, the likely fact of the matter is that super-high res monitors plus a super-powerful graphics card setup remain the preserve of people who have a ton of money not devoted to the true necessities of life.
Most people probably use monitors that have resolutions ranging from 1280 x 1024 to maybe 1680 x 1050. 24" monitors supporting 1920 x 1080 (or even 1920 x 1200) are steadily becoming more affordable, but at the moment they are still rather pricey bar a few exceptions. By pricey, I specifically mean $400.00 or thereabouts and above. This puts them at a price point that is 1 1/2 times more expensive than a corresponding 1680 x 1050 monitor. The question of whether or not such products are a good purchase value is still firmly on the negative side, as far as I'm concerned.
Despite the big divide in pricing between low-end to low-middle-end (1280 x 1024 to 1680 x 1050) and mid-end to high-end (1920 x 1080 to anything higher), at least the monitor market (now dominated by LCDs) has a good history of the products retaining their purchase value. The same cannot be said about the other end of the graphical side of computing, i.e., graphics cards.
In my opinion, the graphics card market is less stable than a particular talent-deprived "pop princess" with delusions of grandeur. The value of products in this particular sector sinks faster than a battleship in quicksand, especially when news and rumors of "the next greatest ever video card" arouses the enthusiast sector's collective tech libido.
Then there's this whole matter of multiple-GPU platforms, which practically defines the motherboard chipset market for the enthusiast sector. Crossfire or SLI? What flavor is your pleasure?
Due to the vicissitudes to which the value in graphics cards are subject and the paucity of affordable high-resolution displays for the common enthusiast (now there's an oxymoron if I ever saw one), it is my firm belief that, given the enthusiast-class GPUs available today, multi-GPU solutions are completely unnecessary. In most cases, a single high-end video card (think in terms of the current champions available today: ATi's RV770 -- either singly or in X2 form-- and nVidia's GT280) is enough to drive big displays with enough power to have the eye candy on and to have playable frame rates. Only monstrously large monitors with insanely high resolutions (think 2560 x 1600 or thereabouts) would cause these video cards to struggle delivering the punch necessary for playable frame rates and max details on.
The hardware manufacturers would all have you believe that, to truly enjoy gaming, you need all the eye candy; you need super-high resolutions; you need more FPS than your mind and eyes can process anyway. (And they sometimes ignore the most basic need of all: You need a good game most of all. But that is a tangential discussion for another time.)
The power of the various hardware providers' marketing divisions is such that you have a gross imbalance between the capabilities of the available hardware (more is better) and the requirements of the software available. Most software (games, specifically) lags behind the curve, so the hardware isn't really used to its maximum.
Then you have the ultimate temptation: Multi-GPU solutions promise a significant increase in performance. In reality, though, because of the limitations of the software (both in terms of drivers and in terms of optimizations of games), the actual performance increases are far short of the monetary investment required. In other words, where Crossfire or SLI requires a 100% (or, with Tri-SLI, a 200%) increase in the cost of the video card, you do not get nearly the same return in performance. The best you can hope for is something akin to 35-40% in the best case scenario. This estimate, by the way, represents the most optimistic appraisal. No matter how you shake it, though, to me that simply is just a very poor return for my investment.
Moreover, for multi-GPU setups to make the most sense, you have to drive a huge resolution, which means a commensurate investment in your display. Today's iterations of Crossfire and SLI are really only for mega-high resolutions; people who own monitors with low-to-average resolutions need not apply.
The power of marketing, however, is so prevalent that many have been led to believe that life is empty without a gigantic resolution and a multi-powerful graphics setup.
I don't know about you, but I personally don't have very deep pockets. I pay for rent, fuel, food, and other necessities. I can't devote around a thousand dollars (or more) on a build just for the graphics setup.
My position is that you don't really need such powerful graphics setups or a high-res monitor to enjoy today's video games. After all, there's more to a good game than just pretty pixels dancing behind a piece of filament. Especially given today's economic realities, we all need to break free from the seductions of marketing.