Benchmark: toFixed vs toPrecision vs Math.round() vs customRound()

Script Preparation code:

var someFloat = 0.123456789;
var customRound = x => (x + (x > 0 ? 0.5 : -0.5)) << 0;

Tests:

toFixed(2)
someFloat.toFixed(2);
toPrecision(2).toString()
someFloat.toPrecision(2).toString();
(Math.round(*100)/100).toString()
(Math.round(someFloat*100)/100).toString();
(Math.round(*100) * 0.01).toString()
(Math.round(someFloat*100)*0.01).toString();
customRound(*100) * 0.01.toString()
(customRound(someFloat*100)*0.01).toString();

Rendered benchmark preparation results:

Suite status: <idle, ready to run>

Previous results

Test case name	Result
toFixed(2)
toPrecision(2).toString()
(Math.round(*100)/100).toString()
(Math.round(100) 0.01).toString()
customRound(100) 0.01.toString()

Fastest: N/A

Slowest: N/A

Latest run results:

Run details: (Test run date: 3 years ago)

User agent: Mozilla/5.0 (Macintosh; Intel Mac OS X 10_15_7) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/94.0.4606.81 Safari/537.36 Edg/94.0.992.50

Browser/OS: Chrome 94 on Mac OS X 10.15.7

View result in a separate tab

Test name	Executions per second
toFixed(2)	3158512.8 Ops/sec
toPrecision(2).toString()	4505558.5 Ops/sec
(Math.round(*100)/100).toString()	5027852.0 Ops/sec
(Math.round(100) 0.01).toString()	5005704.5 Ops/sec
customRound(100) 0.01.toString()	6968986.0 Ops/sec

Autogenerated LLM Summary (model llama3.2:3b, generated 2 months ago):

LLMs can make mistakes. Check important info.

Let's break down what's being tested on MeasureThat.net.

**Benchmark Definition**

The benchmark measures the performance of four different approaches to round a floating-point number:

1. `toFixed(2)`: Formats the float as a fixed-point number with two decimal places.
2. `toPrecision(2).toString()`: Converts the float to a string with 2 precision digits using `toPrecision`.
3. `(Math.round(*100)/100).toString()`: Rounds the float to two decimal places by multiplying by 100, rounding, and then dividing by 100.
4. `(Math.round(*100) * 0.01).toString()`: Rounds the float to one decimal place by multiplying by 100, rounding, and then multiplying by 0.01.

**Library:**

The custom `round` function is defined as:
```javascript
var customRound = x => (x + (x > 0 ? 0.5 : -0.5)) << 0;
```
This function uses bitwise shift to round the float to an integer. The logic is similar to the IEEE 754 standard's rounding behavior, but implemented in a more concise way.

**Pros and Cons:**

1. `toFixed(2)`:
	* Pros: Simple and straightforward implementation.
	* Cons: May not perform well on very large numbers due to string formatting overhead.
2. `toPrecision(2).toString()`:
	* Pros: More flexible than `toFixed`, as it can handle more precision digits.
	* Cons: May be slower due to the conversion to a string and back.
3. `(Math.round(*100)/100).toString()`:
	* Pros: Simple and efficient implementation for rounding to two decimal places.
	* Cons: May not work well for negative numbers or very large numbers.
4. `(Math.round(*100) * 0.01).toString()`:
	* Pros: Simple and efficient implementation for rounding to one decimal place.
	* Cons: May have slight precision issues due to the multiplication by 100.

**Considerations:**

The benchmark primarily focuses on the performance of these four approaches, but it's also worth noting that `toFixed` can lead to issues with NaN (Not a Number) values or very large numbers, while `toPrecision` and the custom rounding function may exhibit better behavior in these cases.

**Alternatives:**

Other rounding functions available in JavaScript include:

1. `Number.EPSILON`: Returns a small value close to zero, which can be used for relative rounding.
2. `Math.floor()` or `Math.ceil()`: Can be used for floor or ceiling rounding, respectively.
3. `decimal.js` library: Provides a more robust decimal arithmetic system with precise control over rounding behavior.

These alternatives may offer better performance or functionality in certain cases, but they also introduce additional complexity and dependencies.

Related benchmarks:

toFixed vs toPrecision vs Math.round() vs customRound() (version: 0)

Comparing performance of: toFixed(2) vs toPrecision(2).toString() vs (Math.round(100)/100).toString() vs (Math.round(100) * 0.01).toString() vs customRound(100) 0.01.toString()

Created: 3 years ago by: Guest

Jump to the latest result

toFixed(2)

toPrecision(2).toString()

(Math.round(*100)/100).toString()

(Math.round(100) 0.01).toString()

customRound(100) 0.01.toString()

Suite status: <idle, ready to run>

Experimental features:

Fastest: N/A

Slowest: N/A

Autogenerated LLM Summary (model llama3.2:3b, generated 2 months ago):

toFixed vs toPrecision vs Math.round() vs customRound() (version: 0)

Comparing performance of: toFixed(2) vs toPrecision(2).toString() vs (Math.round(*100)/100).toString() vs (Math.round(*100) * 0.01).toString() vs customRound(*100) * 0.01.toString()

Created: 3 years ago by: Guest

Jump to the latest result

toFixed(2)

toPrecision(2).toString()

(Math.round(*100)/100).toString()

(Math.round(*100) * 0.01).toString()

customRound(*100) * 0.01.toString()

Suite status: <idle, ready to run>

Experimental features:

Fastest: N/A

Slowest: N/A

Autogenerated LLM Summary (model llama3.2:3b, generated 2 months ago):

Comparing performance of: toFixed(2) vs toPrecision(2).toString() vs (Math.round(100)/100).toString() vs (Math.round(100) * 0.01).toString() vs customRound(100) 0.01.toString()

(Math.round(100) 0.01).toString()

customRound(100) 0.01.toString()