How powerful is most powerful computer possible?

(Note:  this article requires a knowledge of algebra)

As our computer's get faster and smaller, the obvious question to ask is, "How far can our computers go?"  To understand the answer to this problem, we must understand what limits the power of the idealistic computer.  The limits on our computer are the mass/energy of the system, Heisenberg's Uncertainty Principle, and the speed of light travelling through the device.

The formula for the uncertainty principle and energy mass equivalence (to convert between mass and energy) is as follows

  ∆E*∆t ≥  h/4π and E^2=m^2* c^4+ p^2*c^2
   
 We know that to maximize the energy of the system, we want to have our computer flying around at as close to the speed of light as possible, so we can just change the equation to an inequality, and since p = mass * velocity, we have velocity = c

E^2<2* m^2* c^4

We plug in this value into the uncertainty formula

∆t ≥  h/(4π* √2*m*c^2 )

Now, we also must remember to time, we must add the time it takes to send information from one side to the other.  If we call the length of the computer L, we find the additional time to be L/c.  Can L be as small as we want?  Well, if the length get too small, we will get a black hole, and to calculate when this will happen, we use the Chandrashekar limit.

L > GM / c<sup>2

Divide both sides by c to get the additional time.</sup>

L/c > GM / c^3

We add this to our time equation

∆t ≥  h/(4π* √2*m*c^2 ) + GM / c^3

We want to minimize this, so we take the derivative of this function to find where the slope of the function is zero, which will be the minimum time.  (Note:  since we want the ideal machine, we will change the inequality to an equality, which will make this an approximation.)

d∆t/dm =  -h/(4π* √2*m^2*c^2 ) + G / c^3

h/(4π* √2*m^2*c^2 ) = G / c^3

(h * c^3) / (G * 4π* √2*c^2) = (hc)/(G * 4π* √2) ~ 1.6748505 * 10 ^-16
Therefore, the maximum bits per second computable is one over our answer, which is 

6.0* 10 ^15 bits per second

So, how long would it take to "solve" chess with our ideal computer?  An article by chess.com 
(read here: http://www.chess.com/blog/watcha/limits-of-quantum-computing-in-solving-chess)
has suggested that it would take up to 10^42 bits, so to find the amount of time it takes for our 
computer to find the answer, we divide the bits by the rate

10^42 bits / 6.0 * 10 ^ 15 bits per second = 1.7 * 10 ^ 26 seconds, which to put in perspective,
the age of the universe is 4.35 +- .02* 10^17 according to most scientists.  Our best computer 
actually isn't all that fast.