Benchmark: fast absolute value (branched & branchless) vs abs vs ** -1

Script Preparation code:

var abs = Math.abs;
var fastAV = a => a - ((a < 0) * a * 2);
var fastAV2 = a => a < 0 ? -a : a;
var fastAV3 = a => (-1) ** (a > 0) * a;

Tests:

native
abs(-10); abs(10); abs(-10.1); abs(10.1);
custom (branchless)
fastAV(-10); fastAV(10); fastAV(-10.1); fastAV(10.1);
custom (branched)
fastAV2(-10); fastAV2(10); fastAV2(-10.1); fastAV2(10.1);
-1 ** a
fastAV3(-10); fastAV3(10); fastAV3(-10.1); fastAV3(10.1);

Rendered benchmark preparation results:

Suite status: <idle, ready to run>

Previous results

Test case name	Result
native
custom (branchless)
custom (branched)
-1 ** a

Fastest: N/A

Slowest: N/A

Latest run results:

Run details: (Test run date: one year ago)

User agent: Mozilla/5.0 (Macintosh; Intel Mac OS X 10_15_7) AppleWebKit/605.1.15 (KHTML, like Gecko) Version/17.6 Safari/605.1.15

Browser/OS: Safari 17 on Mac OS X 10.15.7

View result in a separate tab

Test name	Executions per second
native	8572205.0 Ops/sec
custom (branchless)	8792451.0 Ops/sec
custom (branched)	9023908.0 Ops/sec
-1 ** a	8547974.0 Ops/sec

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

LLMs can make mistakes. Check important info.

Let's dive into the world of JavaScript microbenchmarks on MeasureThat.net.

**Benchmark Definition**

The benchmark measures the performance of three different approaches to calculate the absolute value (abs) of a number: `Math.abs`, a branchless approach (`fastAV`), and two branched approaches (`fastAV2` and `fastAV3`). The latter two use special JavaScript syntax, `-1 ** a`.

**Approaches Compared**

The benchmark compares four test cases:

1. **Native**: Uses the built-in `Math.abs` function.
2. **Custom (Branchless)**: Implements an optimized branchless approach using bitwise operations (`fastAV`).
3. **Custom (Branched)**: Uses two different branched approaches with special JavaScript syntax (`fastAV2` and `fastAV3`).

**Pros and Cons of Each Approach**

1. **Native**: Pros:
	* Easy to implement and understand.
	* Well-optimized by the browser.
	Cons:
	* May have performance overhead due to function call overhead.
2. **Custom (Branchless)**: Pros:
	* Optimized for performance using bitwise operations.
	* Avoids function call overhead.
Cons:
	* Requires knowledge of bitwise operations and their implications on performance.
3. **Custom (Branched)**: Pros:
	* Easy to implement and understand, especially for developers familiar with the syntax.
	Cons:
	* May have performance overhead due to branching.
4. **-1 ** a**: Pros:
	* Unconventional approach using special JavaScript syntax.
	Cons:
	* May not be easily understandable or maintainable.

**Library Used**

None explicitly mentioned in the benchmark definition. However, `Math.abs` is a built-in function provided by JavaScript engines, including Safari.

**Special JS Feature or Syntax**

The `-1 ** a` approach uses a special JavaScript syntax known as "exponentiation" with a negative base and fractional exponent. This syntax is not widely supported and may not work in all environments.

**Other Alternatives**

If you want to explore more performance-critical optimizations, you can consider using:

* SIMD (Single Instruction, Multiple Data) instructions, which are available on modern CPUs.
* Just-In-Time (JIT) compilation, which can optimize JavaScript code at runtime.

Keep in mind that these alternatives require a good understanding of computer science and programming languages.

I hope this explanation helps! Let me know if you have any further questions.

Related benchmarks:

fast absolute value (branched & branchless) vs abs vs ** -1 (version: 0)

Comparing performance of: native vs custom (branchless) vs custom (branched) vs -1 ** a

Created: 3 years ago by: Guest

Jump to the latest result

native

custom (branchless)

custom (branched)

-1 ** a

Suite status: <idle, ready to run>

Fastest: N/A

Slowest: N/A

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