A Matrix/Vector library for elm. Up to 4 dimensions. 3D Transformations mostly.
Home Page: http://package.elm-lang.org/packages/Zinggi/elm-webgl-math/latest
License: MIT License
Please run the benchmark here:
https://zinggi.github.io/elm-webgl-math/
And post your result here.
Please use this tool to generate a nice markdown.
To get a nice output for chrome mobile I used USB remote debugging to copy the html to here. If you know an easier way, please let me know!
Here are the results from my laptop:
| suite | benchmark | ops/sec | error% | samples |
|---|---|---|---|---|
| Matrix exponentiation: A^1000 | M4.pow | 5699 | 10.7 | 16 |
| Matrix exponentiation: A^1000 | RefM4.pow | 6310 | 8.2 | 27 |
| Matrix exponentiation: A^10000 | M4.pow | 645 | 6.0 | 26 |
| Matrix exponentiation: A^10000 | RefM4.pow | 662 | 3.1 | 26 |
| Matrix exponentiation: A^100000 | M4.pow | 66 | 3.7 | 23 |
| Matrix exponentiation: A^100000 | RefM4.pow | 70 | 1.8 | 23 |
| Matrix exponentiation: A^1000000 | M4.pow | 7 | 2.0 | 11 |
| Matrix exponentiation: A^1000000 | RefM4.pow | 7 | 1.4 | 11 |
| Useless vector addition/scale: f a b = f (a+b) (-0.999999 * b), 1000 times. | V4.add / scale | 14378 | 2.2 | 27 |
| Useless vector addition/scale: f a b = f (a+b) (-0.999999 * b), 1000 times. | RefV4.add / scale | 5386 | 4.5 | 15 |
| Useless vector addition/scale: f a b = f (a+b) (-0.999999 * b), 10000 times. | V4.add / scale | 1345 | 8.6 | 25 |
| Useless vector addition/scale: f a b = f (a+b) (-0.999999 * b), 10000 times. | RefV4.add / scale | 566 | 2.4 | 19 |
| Useless vector addition/scale: f a b = f (a+b) (-0.999999 * b), 100000 times. | V4.add / scale | 137 | 7.3 | 25 |
| Useless vector addition/scale: f a b = f (a+b) (-0.999999 * b), 100000 times. | RefV4.add / scale | 56 | 3.4 | 22 |
| Real world application: makeTransform 1000 times. | M4.makeTransform | 1894 | 5.3 | 25 |
| Real world application: makeTransform 1000 times. | M4.makeTransformNaive | 1277 | 8.7 | 25 |
| Real world application: makeTransform 1000 times. | RefM4.makeTransform | 754 | 5.9 | 27 |
| Real world application: makeTransform 10000 times. | M4.makeTransform | 195 | 5.2 | 27 |
| Real world application: makeTransform 10000 times. | M4.makeTransformNaive | 126 | 7.4 | 24 |
| Real world application: makeTransform 10000 times. | RefM4.makeTransform | 79 | 1.6 | 23 |
| Real world application: makeTransform 100000 times. | M4.makeTransform | 19 | 8.8 | 17 |
| Real world application: makeTransform 100000 times. | M4.makeTransformNaive | 14 | 2.5 | 13 |
| Real world application: makeTransform 100000 times. | RefM4.makeTransform | 8 | 2.1 | 12 |
| Rotate vector around 1000 times | M4.transform | 13050 | 6.5 | 23 |
| Rotate vector around 1000 times | RefM4.transform | 6357 | 2.0 | 17 |
| Rotate vector around 10000 times | M4.transform | 1446 | 1.9 | 26 |
| Rotate vector around 10000 times | RefM4.transform | 591 | 8.3 | 26 |
| Rotate vector around 100000 times | M4.transform | 123 | 12.8 | 23 |
| Rotate vector around 100000 times | RefM4.transform | 63 | 2.7 | 22 |
| suite | benchmark | ops/sec | error% | samples |
|---|---|---|---|---|
| Matrix exponentiation: A^1000 | M4.pow | 8659 | 1.9 | 27 |
| Matrix exponentiation: A^1000 | RefM4.pow | 2501 | 1.3 | 28 |
| Matrix exponentiation: A^10000 | M4.pow | 758 | 2.1 | 27 |
| Matrix exponentiation: A^10000 | RefM4.pow | 253 | 2.6 | 25 |
| Matrix exponentiation: A^100000 | M4.pow | 99 | 2.0 | 23 |
| Matrix exponentiation: A^100000 | RefM4.pow | 27 | 2.0 | 18 |
| Matrix exponentiation: A^1000000 | M4.pow | 10 | 1.9 | 14 |
| Matrix exponentiation: A^1000000 | RefM4.pow | 3 | 1.4 | 7 |
| Useless vector addition/scale: f a b = f (a+b) (-0.999999 * b), 1000 times. | V4.add / scale | 16119 | 3.2 | 26 |
| Useless vector addition/scale: f a b = f (a+b) (-0.999999 * b), 1000 times. | RefV4.add / scale | 1433 | 6.3 | 24 |
| Useless vector addition/scale: f a b = f (a+b) (-0.999999 * b), 10000 times. | V4.add / scale | 1567 | 2.8 | 27 |
| Useless vector addition/scale: f a b = f (a+b) (-0.999999 * b), 10000 times. | RefV4.add / scale | 154 | 4.7 | 25 |
| Useless vector addition/scale: f a b = f (a+b) (-0.999999 * b), 100000 times. | V4.add / scale | 154 | 5.0 | 25 |
| Useless vector addition/scale: f a b = f (a+b) (-0.999999 * b), 100000 times. | RefV4.add / scale | 17 | 1.2 | 15 |
| Real world application: makeTransform 1000 times. | M4.makeTransform | 5876 | 9.6 | 24 |
| Real world application: makeTransform 1000 times. | M4.makeTransformNaive | 2090 | 2.5 | 24 |
| Real world application: makeTransform 1000 times. | RefM4.makeTransform | 186 | 6.2 | 23 |
| Real world application: makeTransform 10000 times. | M4.makeTransform | 560 | 8.6 | 21 |
| Real world application: makeTransform 10000 times. | M4.makeTransformNaive | 208 | 3.7 | 25 |
| Real world application: makeTransform 10000 times. | RefM4.makeTransform | 21 | 2.5 | 18 |
| Real world application: makeTransform 100000 times. | M4.makeTransform | 63 | 2.3 | 21 |
| Real world application: makeTransform 100000 times. | M4.makeTransformNaive | 17 | 21.5 | 15 |
| Real world application: makeTransform 100000 times. | RefM4.makeTransform | 2 | 4.3 | 6 |
| Rotate vector around 1000 times | M4.transform | 7523 | 7.1 | 25 |
| Rotate vector around 1000 times | RefM4.transform | 2592 | 8.3 | 27 |
| Rotate vector around 10000 times | M4.transform | 773 | 3.3 | 24 |
| Rotate vector around 10000 times | RefM4.transform | 272 | 6.4 | 25 |
| Rotate vector around 100000 times | M4.transform | 69 | 8.4 | 21 |
| Rotate vector around 100000 times | RefM4.transform | 28 | 4.2 | 18 |
| suite | benchmark | ops/sec | error% | samples |
|---|---|---|---|---|
| Matrix exponentiation: A^1000 | M4.pow | 1150 | 10.2 | 18 |
| Matrix exponentiation: A^1000 | RefM4.pow | 1129 | 6.1 | 21 |
| Matrix exponentiation: A^10000 | M4.pow | 125 | 7.3 | 21 |
| Matrix exponentiation: A^10000 | RefM4.pow | 119 | 7.9 | 22 |
| Matrix exponentiation: A^100000 | M4.pow | 12 | 19.8 | 12 |
| Matrix exponentiation: A^100000 | RefM4.pow | 12 | 11.1 | 15 |
| Matrix exponentiation: A^1000000 | M4.pow | 1 | 5.4 | 6 |
| Matrix exponentiation: A^1000000 | RefM4.pow | 1 | 9.3 | 6 |
| Useless vector addition/scale: f a b = f (a+b) (-0.999999 * b), 1000 times. | V4.add / scale | 2716 | 11.3 | 24 |
| Useless vector addition/scale: f a b = f (a+b) (-0.999999 * b), 1000 times. | RefV4.add / scale | 799 | 14.7 | 21 |
| Useless vector addition/scale: f a b = f (a+b) (-0.999999 * b), 10000 times. | V4.add / scale | 264 | 8.9 | 21 |
| Useless vector addition/scale: f a b = f (a+b) (-0.999999 * b), 10000 times. | RefV4.add / scale | 91 | 6.7 | 20 |
| Useless vector addition/scale: f a b = f (a+b) (-0.999999 * b), 100000 times. | V4.add / scale | 30 | 2.8 | 17 |
| Useless vector addition/scale: f a b = f (a+b) (-0.999999 * b), 100000 times. | RefV4.add / scale | 9 | 7.1 | 12 |
| Real world application: makeTransform 1000 times. | M4.makeTransform | 333 | 9.4 | 22 |
| Real world application: makeTransform 1000 times. | M4.makeTransformNaive | 132 | 13.3 | 19 |
| Real world application: makeTransform 1000 times. | RefM4.makeTransform | 112 | 5.9 | 21 |
| Real world application: makeTransform 10000 times. | M4.makeTransform | 37 | 3.0 | 19 |
| Real world application: makeTransform 10000 times. | M4.makeTransformNaive | 13 | 32.2 | 12 |
| Real world application: makeTransform 10000 times. | RefM4.makeTransform | 10 | 11.9 | 14 |
| Real world application: makeTransform 100000 times. | M4.makeTransform | 4 | 5.8 | 8 |
| Real world application: makeTransform 100000 times. | M4.makeTransformNaive | 2 | 12.6 | 6 |
| Real world application: makeTransform 100000 times. | RefM4.makeTransform | 1 | 13.2 | 6 |
| Rotate vector around 1000 times | M4.transform | 2659 | 6.5 | 9 |
| Rotate vector around 1000 times | RefM4.transform | 1204 | 22.9 | 21 |
| Rotate vector around 10000 times | M4.transform | 245 | 9.3 | 21 |
| Rotate vector around 10000 times | RefM4.transform | 119 | 12.0 | 20 |
| Rotate vector around 100000 times | M4.transform | 27 | 4.7 | 16 |
| Rotate vector around 100000 times | RefM4.transform | 12 | 13.2 | 12 |
First: library is very interesting and would be directly useful for a couple of projects of mine, so thank you.
Followed the instructions at https://github.com/Zinggi/elm-linear-algebra/tree/master/bench
nw@salad ~/elm/elm-linear-algebra/bench master$ git clone https://github.com/fredcy/elm-benchmark
Cloning into 'elm-benchmark'...
remote: Counting objects: 483, done.
remote: Total 483 (delta 0), reused 0 (delta 0), pack-reused 483
Receiving objects: 100% (483/483), 266.72 KiB | 202.00 KiB/s, done.
Resolving deltas: 100% (161/161), done.
Checking connectivity... done.
nw@salad ~/elm/elm-linear-algebra/bench master$ elm package install --yes
Starting downloads...
● elm-lang/virtual-dom 1.1.1
● elm-lang/html 1.1.0
● elm-community/elm-linear-algebra 2.0.3
● elm-lang/core 4.0.5
● ggb/numeral-elm 1.2.2
Packages configured successfully!
nw@salad ~/elm/elm-linear-algebra/bench master$ elm make Main.elm --output elm.js
Success! Compiled 50 modules.
Successfully generated elm.js
nw@salad ~/elm/elm-linear-algebra/bench master$ chromium-browser index.html
Created new window in existing browser session.
When I switch to Chromium I get "See javascript console if no output appears below."
The console sez:
file://cdn.jsdelivr.net/pure/0.6.0/pure-min.css Failed to load resource: net::ERR_FILE_NOT_FOUND
file://cdn.jsdelivr.net/lodash/4.13.1/lodash.min.js Failed to load resource: net::ERR_FILE_NOT_FOUND
file://cdn.jsdelivr.net/platform.js/1.3.1/platform.js Failed to load resource: net::ERR_FILE_NOT_FOUND
file://cdn.jsdelivr.net/benchmarkjs/2.1.0/benchmark.js Failed to load resource: net::ERR_FILE_NOT_FOUND
elm.js:11114 Uncaught ReferenceError: Benchmark is not defined
index.html:23 Uncaught ReferenceError: Elm is not defined
Suggestion: why don't you just create a gh-pages branch with the benchmark built & deployed, so that users can just click there and give you the benchmark results, without any clone/build shenanigans?
Hi, the license info in elm-package.json (BSD3) and LICENSE (MIT) are different. Possible to ensure these are the same?
React
A declarative, efficient, and flexible JavaScript library for building user interfaces.
🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.
Typescript
TypeScript is a superset of JavaScript that compiles to clean JavaScript output.
An Open Source Machine Learning Framework for Everyone
Django
The Web framework for perfectionists with deadlines.
Laravel
A PHP framework for web artisans
Bring data to life with SVG, Canvas and HTML. 📊📈🎉
JavaScript (JS) is a lightweight interpreted programming language with first-class functions.
Some thing interesting about web. New door for the world.
A server is a program made to process requests and deliver data to clients.
Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.
Some thing interesting about visualization, use data art
Some thing interesting about game, make everyone happy.
Facebook
We are working to build community through open source technology. NB: members must have two-factor auth.
Microsoft
Open source projects and samples from Microsoft.
Google
Google ❤️ Open Source for everyone.
Alibaba
Alibaba Open Source for everyone
D3
Data-Driven Documents codes.
Tencent
China tencent open source team.