The mistake you're making here is failing to understand the difference between constant factor and asymptotic bounds. In computer science, algorithm analysis explicitly ignores constant factors like computer speed, compiler speed, language speed etc... because asymptotically they are irrelevant.
For non math geeks, what I mean by asymptotic bounds is the curve of execution time as the number of inputs increases. When the number of inputs is low, the dominating concern is the constant factor, but as inputs increase this quickly becomes irrelevant.
Consider this: let's calculate every possible permutation of a set of inputs. Starting with two inputs, [1,2] then it's easy - { [1,2], [2,1] }. Still pretty easy when we go to three inputs: { [1,2,3], [1,3,2], [2,1,3],[2,3,1],[3,2,1],[3,1,2] }, but when we go to four it starts to become more annoying, and by the time we go to 15 or 20, we're getting crazy with the possible number of combinations. That's because the number of possible permutations is defined by the factorial n! (a factorial is numbers multiplied in decreasing order, so 5! means 5 * 4 * 3 * 2 * 1).
Therefore calculating the possible number of permutations of a few numbers is easy, but calculating the number of permutations for many numbers takes a long time. It doesn't matter how fast the computer is, I can simply add more numbers and I will quickly outstrip your ability to calculate it.
In fact, this is going to be pretty much true of anything that runs in higher order time ( O(n ^ x) ) because as the number of inputs increases, the time required to calculate it increases exponentially. As long as x > 1, eventually, no matter how fast your computer is, the constant factor improvements gained by your computer speed will be dominated by the exponent x.
Now, when you start talking about P and NP, things get more complex, but for simplicity's sake imagine that the class of algorithms that are P are things we can "put our finger on", or technically, those whose execution times are bounded by a polynomial. NP problems are even more complex than this, in that they can't even be bounded within polynomial time constraints. Factoring numbers is part of the NP problem space, and cryptography relies on this.
The exciting/terrifying thing about this news is the notion that we can take a problem from the NP class and make it P, and if this is the case it has profound implications.