Questions tagged [floating-point]

Mathematical questions concerning floating point numbers, a finite approximation of the real numbers used in computing.

472 questions
77
votes
4 answers

Why does the google calculator give $\tan 90^{\circ} = 1.6331779e^{+16}$?

I typed in $\tan 90^{\circ}$ in Google and it gave $1.6331779\mathrm{E}16$. How did it come to this answer? Limits? Some magic?
Gizmo
  • 939
58
votes
1 answer

Show that floating point $\sqrt{x \cdot x} \geq x$ for all long $x$.

I verified experimentally that in Java the equality Math.sqrt(x*x) = x holds for all long x such that x*x doesn't overflow. Here, Java long is a $64$ bit signed type and double is a IEEE binary floating point type with at least $53$ bits mantissa…
maaartinus
  • 1,401
25
votes
5 answers

Accurate floating-point linear interpolation

I want to perform a simple linear interpolation between $A$ and $B$ (which are binary floating-point values) using floating-point math with IEEE-754 round-to-nearest-or-even rounding rules, as accurately as possible. Please note that speed is not a…
Pedro Gimeno
  • 773
  • 8
  • 13
21
votes
1 answer

Why is 'catastrophic cancellation' called so?

I was studying Numerical Analysis by K. Mukherjee; there he discussed Loss of Significant Figures by Subtraction, as followed: In the subtraction of two approximate numbers, a serious type of error may be present when the numbers are nearly equal.…
user142971
20
votes
3 answers

Polynomial for very large number of roots

Update: Please also see this solution here provided by MattL. I have the roots for a very large order polynomial (>100), and from those alone wish to recreate the polynomial and run into numerical challenges when using the poly function for doing…
16
votes
5 answers

Solving a quadratic equation with precision when using floating point variables

I know how to solve a basic quadratic equation with the formula $$t_{1,2}=\dfrac{-b\pm\sqrt{b^2-4ac}}{2a}$$ but I learned that if $b \approx \sqrt{b^2-4ac}$ floating point precision may give slightly wrong results and this approach is better. It…
16
votes
1 answer

Explain why catastrophic cancellation happens

After my own research, the following picture emerges as the most frequently used example of catastrophic cancellation (It is indeed used in my class). Could anyone explain why the plot takes that shape? (i.e., the widely fluctuating jagged line…
15
votes
3 answers

Plotting $\left(1+\frac{1}{x^n}\right)^{x^n}$.

When I plot the following function, the graph behaves strangely: $$f(x) = \left(1+\frac{1}{x^{16}}\right)^{x^{16}}$$ While $\lim_{x\to +\infty} f(x) = e$ the graph starts to fade at $x \approx 6$. What's going on here? (plotted on my trusty old 32…
14
votes
2 answers

Algebraic structure for the floating-point arithmetic.

Is it possible to consider floating-point arithmetic as some common algebraic structure? For example, consider something like simplified IEEE754 single precision binary floating-point subset of numbers that consist of 1 sign bit, 23-bit mantissa…
14
votes
2 answers

Moving point along the vector

I'm making a video game in which a ball is moving towards a player. So, I have a point $P$ describing the point where the player is, and a point $B$ describing where the ball is. I know that we can represent the direction from $B$ to $A$ as a…
13
votes
1 answer

Why is the maximum value of my calculator $10^{100}$?

The maximum value I can have in my calculator is $10^{100}$. At first it seems right but when we consider that the calculator should stores its number as binary, this seems weird. Why isn't the maximum exponent a base 2 number? Is this limit set…
Winter
  • 946
12
votes
1 answer

What's special about the number $1.000000015047466$E+$30$?

I'm a programmer by trade by I've run into a weirdly special number and need some help deciphering its significance. I was writing some machine learning code that compiles into GPU kernel code and the compiler output the number 1.000000015047466E+30…
10
votes
2 answers

How does rewriting $x^2 -y^2$ as $(x+y)(x-y)$ avoid catastrophic cancellation?

Why is rewriting $x^2 -y^2$ as $(x+y)(x-y)$ a way to avoid catastrophic cancellation? We are still doing $(x-y)$. Is it because the last operation in the second form is a multiplication?
9
votes
2 answers

Mathematically, why does this Python program give such an inaccurate result for the Taylor series of $x \mapsto \exp\left(x\right)$ at $x=-40$?

from math import pow, factorial if __name__ == '__main__': # Parameters x = -40 n_terms = 124 sum = 0 for i in range(124): sum += pow(x, i) / factorial(i) print(sum) Output: 5.881161314775606 We know $e^{-40}$ is…
9
votes
2 answers

Conditioning of the linear systems in the inverse or Rayleigh quotient iteration algorithms

I'm working through the book Numerical Linear Algebra by Trefethen and Bau. In Lecture 27 (and exercise 27.5), the following claim is made about the inverse iteration algorithm: Let $ A $ be a real, symmetric matrix. Solving the system $ (A - \mu…
1
2 3
31 32