Stack Exchange

njuffa

top accounts reputation activity bookmarks subscriptions
Top Questions
47 votes

Efficient floating-point division with constant integer divisors

c algorithm math floating-point division asked Feb 20 '16 at 19:01
stackoverflow.com

30 votes

Optimized low-accuracy approximation to `rootn(x, n)`

algorithm math floating-point bit-manipulation asked Aug 17 '15 at 4:10
stackoverflow.com

25 votes

Efficient computation of the average of three unsigned integers (without overflow)

c algorithm bit-manipulation micro-optimization extended-precision asked Oct 27 '20 at 21:21
stackoverflow.com

24 votes

Best machine-optimized polynomial minimax approximation to arctangent on [-1,1]?

c algorithm floating-point asked Nov 1 '14 at 20:22
stackoverflow.com

21 votes

Efficiently computing (a - K) / (a + K) with improved accuracy

c algorithm floating-point floating-accuracy asked Feb 16 '16 at 4:51
stackoverflow.com

17 votes

Efficient computation of 2**64 / divisor via fast floating-point reciprocal

c floating-point bit-manipulation division asked Apr 26 '16 at 1:28
stackoverflow.com

14 votes

Optimal 9-element sorting network that reduces to an optimal median-of-9 network?

algorithm median sorting-network asked Aug 2 '17 at 6:54
stackoverflow.com

13 votes

Implementation of sinpi() and cospi() using standard C math library

c floating-point trigonometry math.h asked Mar 14 '17 at 17:39
stackoverflow.com

8 votes

What is the origin of “home free”?

etymology idioms asked Jul 15 '18 at 18:15
english.stackexchange.com

8 votes

Accurate and efficient computation of the inverse Langevin function

algorithms efficiency accuracy special-functions asked Sep 21 '18 at 21:30
scicomp.stackexchange.com

1 2

Top Answers
88
Emulate “double” using 2 “float”s
stackoverflow.com

66
Why was the caret used for XOR instead of exponentiation?
softwareengineering.stackexchange.com

52
128 bit integer on cuda?
stackoverflow.com

31
Numerically stable way to compute sqrt((b²*c²) / (1-c²)) for c in [-1, 1]
stackoverflow.com

31
What did Pete Stewart think he knew about efficient implementation of floating point denormals?
retrocomputing.stackexchange.com

25
Fastest Implementation of the Natural Exponential Function Using SSE
stackoverflow.com

22
How is arctan implemented?
stackoverflow.com

20
Who was Heinrich Kornblum?
hsm.stackexchange.com

20
Is it possible to put assembly instructions into CUDA code?
stackoverflow.com

20
Algorithm to convert an IEEE 754 double to a string?
stackoverflow.com

19
Why does Math.cbrt(1728) produce a more accurate result than Math.pow(1728, 1/3)?
stackoverflow.com

18
How to find the position of the only-set-bit in a 64-bit value using bit manipulation efficiently?
stackoverflow.com

17
Which x87 first acquired the ability to flush denormal to zero?
retrocomputing.stackexchange.com

17
Why is “a =(b>0)?1:0” better than “if-else” version in CUDA?
stackoverflow.com

17
Why do gcc and NVCC (g++) see two different structure sizes?
stackoverflow.com

17
How do I develop numerical routines for the evaluation of my own special functions?
math.stackexchange.com

16
Numerically stable method for solving quadratic equations
stackoverflow.com

15
Why do computer-science majors learn calculus?
cseducators.stackexchange.com

15
Algorithm for pow(float, float)
stackoverflow.com

14
Very fast approximate Logarithm (natural log) function in C++?
stackoverflow.com

14
x87 trigonometric instructions
retrocomputing.stackexchange.com

13
Units of cuda registers
stackoverflow.com

13
Does this usage of strncmp contain an out of bounds read?
stackoverflow.com

13
Efficient Algorithm for Bit Reversal (from MSB->LSB to LSB->MSB) in C
stackoverflow.com

12
Atan2 Faster Approximation
math.stackexchange.com

12
Why do higher-precision floating point formats have so many exponent bits?
stackoverflow.com

12
Whittaker-Shannon interpolation: Accuracy dies with speedup; can it be fixed?
scicomp.stackexchange.com

12
cuda inline and noinline device functions
stackoverflow.com

12
In what types of loops is it best to use the #pragma unroll directive in CUDA?
stackoverflow.com

12
Compute logarithmic expression without floating point arithmetics or log
stackoverflow.com

1 2 3 4