"Not enough mana!"

1 - 0.9 - 0.1 = 0? Apparently not! (there is this joke: Q: when 2 + 2 = 5? A: for large values of 2)

And speaking of errors in real life...Fake or Photo?

Computers are getting a hard time removing the error in a simple computation, but have a hard time trying to represent imperfect things from real life...

Update:
Seems to me, no one has this wright (and probably never will because of that floating point binary representation)

• Google's Calculator: 1 - 0.9 - 0.1 = -2.77555756 × 10^-17
• Ruby: 1 - 0.9 - 0.1 = -2.77555756156289e-17
• PHP(4.4): 1 - 0.9 - 0.1 = -2.7755575615629e-17
• JavaScript:1 - 0.9 - 0.1 = -2.7755575615628914e-17
• VbScript: 1 - 0.9 - 0.1 = -2,77555756156289E-17
• Java2: 1 - 0.9 - 0.1 = -2.7755575615628914E-17
• Perl 5: 1 - 0.9 - 0.1 = -2.77555756156289e-017
• C#: 1 - 0.9 - 0.1 = -2.7755575615628914E-17
• VC++: 1 - 0.9 - 0.1 = -2.7755575615628914e-017

Instead, 1 - 0.1 - 0.9 = 0, at least in VC++, C# and Google's Calculator...