Current microprocessors are very complex beasts. To develop an high-performance CPU architecture, you not only need many very smart engineers, but much time (3-5 years) and money (in the order of billions $$$). Moreover, bleeding-edge fabrication plants are incredibly expensive, and they must be continuously upgraded to newer process technologies.
So, it is perfectly understandable that both AMD and Intel (the two main x86 players) try to differentiate they offer, selling processors that spans from 50$ to ~1000$, a range of about 20X. While they want to sell you the most high-price (and high-margin) processors, they also realize that, as the market is very cost-sensitive, the bulk of R&D and production costs must be spread over a very large, low-profit product base. On top of that, all their processors must perform at least decently, or user will loudly complain (hello, Atom users!).
This means a somewhat precarious balance: while they had to sell many low cost, yet decently performing processor, they should be careful to not make low cost processors that are too much good for their intended market, otherwise high margin sells will stagnate (as users don't see the need for a faster processor).
So, the key word is differentiate. Traditionally, this differentiation can be carried out in two different manners: by performance and by features.
The first is straightforward: the less you pay, the slower the CPU. However, while in the past this approach worked quite well, current CPUs are so quick that many peoples will never realize that their processor is slower then an high-end one.
Consequently, you had to differentiate by features. Your low-end processor should not be able to process all the instructions that your high-end CPU can, so that the user will "feel" the need to upgrade (or to directly buy and higher specs processor). While missing some features can sometimes means that you can not run a particular software, generally it means that some specific program will run 5-10 times slower, so that the user will complain about its speed (and will buy a pricier CPU).
In a sense, it seems that features capping is not so different that performance limitations, but things is not so simple in real world. As you will see, handicapping CPU features too much can have a big detrimental consequence on the entire x86 ecosystem. It is of no help than Intel, the n.1 x86 CPU vendor (and technology leader), put some inexplicable, intricate and complex features caps. In the past AMD was also guilty (the first Semprons had AMD64 disabled) but lately, forced by Intel performance dominance, they played fair game regarding CPU capabilities.
Lets face it: some differentiation is absolutely necessary. However, in my opinion, Intel really exceeded in this marketing discipline.