# truncb > Round a numeric value to the nearest multiple of b^n toward zero.
## Usage ```javascript var truncb = require( '@stdlib/math/base/special/truncb' ); ``` #### truncb( x, n, b ) Rounds a `numeric` value to the nearest multiple of `b^n` toward zero. ```javascript // Round a value to 4 decimal places: var v = truncb( 3.141592653589793, -4, 10 ); // returns 3.1415 // If n = 0 or b = 1, `truncb` behaves like `trunc`: v = truncb( 3.141592653589793, 0, 2 ); // returns 3.0 // Round a value to the nearest multiple of two toward zero: v = truncb( 5.0, 1, 2 ); // returns 4.0 ```
## Notes - Due to rounding error in [floating-point numbers][ieee754], rounding may **not** be exact.
## Examples ```javascript var randu = require( '@stdlib/random/base/randu' ); var round = require( '@stdlib/math/base/special/round' ); var pow = require( '@stdlib/math/base/special/pow' ); var truncb = require( '@stdlib/math/base/special/truncb' ); var x; var n; var b; var v; var i; for ( i = 0; i < 100; i++ ) { x = (randu()*100.0) - 50.0; n = round( (randu()*10.0) - 5.0 ); b = round( randu()*10.0 ); v = truncb( x, n, b ); console.log( 'x: %d. %d^%d: %d. Rounded: %d.', x, b, n, pow( b, n ), v ); } ```